sessionId=019e87c4-7d2d-7c72-8c85-712a6cfe326f
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/announcement_state.json
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/chat_history.jsonl
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/events.jsonl
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/hunk_records.jsonl
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/prompt_context.json
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/rewind_points.jsonl
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/signals.json
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/summary.json
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/system_prompt.txt
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e87c4-7d2d-7c72-8c85-712a6cfe326f/updates.jsonl
## LazyCodex/status search
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e7466-d18d-71b0-9b1f-6599dc5defbc/updates.jsonl:7695:{"timestamp":1780087109,"method":"session/update","params":{"sessionId":"019e7466-d18d-71b0-9b1f-6599dc5defbc","update":{"sessionUpdate":"tool_call_update","toolCallId":"call-29eb3720-555f-44d3-8b22-2a482d8b3d62-50","status":"completed","content":[{"type":"content","content":{"type":"text","text":"1→# Implementation Plan: OMO-like Harnessing Agent OS on Grok Build (LFE-core + LFG Adaptation Layer + GPT-5.5 Orchestrator)\n2→\n3→**Session:** 019e7466-d18d-71b0-9b1f-6599dc5defbc  \n4→**Date:** 2026-05-30 (revised after user clarification)  \n5→**Status:** Ready for user approval (Plan Mode) — **Broad scope: full LFE-core-powered OMO harnessing Agent OS on Grok Build**\n6→\n7→---\n8→\n9→## Context (User's Clarified Vision — \"in short\")\n10→\n11→**User's direct clarification:** \"in short is this the plan to creat omo like harnessing agent os that runs on grok build with supporting main ochestrator be using gpt5.5 not grok build on grok build with same manner of adaptation layer and hook and featurew like lazy lsp and shared skills and keyword triggering logic with omo and lfe-core must use lfe-core to implment this feature put lfe-core as submodules like lfp\"\n12→\n13→**Yes — this revised plan is exactly that.**\n14→\n15→**MANDATORY FIRST STEP FOR ANY IMPLEMENTER (per user feedback at plan review — including @plan.md:14):** Before writing *any* code, the person executing this plan **must**:\n16→1. `gh pr view 4471 --repo code-yeongyu/oh-my-openagent --json files,commits,body` (or equivalent full diff + discussion) + read *every* related doc and \"omo-\" file touched by the PR.\n17→2. Globally search the LFG codebase (and the omo-standalone vendor) for any \"omo-\" prefixed duplication in the code hierarchy (agent definitions, hook shapes, skill catalogs, LSP/MCP lazy logic, shared skills, keyword triggering, continuation, review synthesis, etc.).\n18→3. Re-read LFP's `docs/lazycodex-reverse-engineering.md`, `lfe-core-integration.md`, `lfe-core-agent-os-consumption.md`, and the actual `lfe-core/src/Lfe.CodexAdapter/` + `Lfe.CodexMcpBridge/` + `Lfe.AgentOs/` implementations.\n19→4. **Decide and document** how LFE-core becomes the single unifying core (removing OMO duplication where LFE-core now provides the neutral contract). The adaptation layer created after PR #4471 learnings must sit on top of LFE-core, not duplicate OMO logic.\n20→5. Re-audit *this plan.md* against the above and update it with the unification strategy before any implementation starts.\n21→6. Only then begin Phase 0.\n22→\n23→This \"check OMO PR 4471 + remove omo- duplication + use lfe-core for unification + check your plan again\" gate is non-negotiable and must be evidenced in the first commit or PR description. LFE-core is the unification path to OMO parity (post-4471), not a second parallel stack.\n24→\n25→**Initial Audit Findings (executed by planner + re-checked per user @plan.md:14 feedback, 2026-05-30):**\n26→- OMO PR 4471 artifacts (from exploration of the changed files in `src/cli/install-codex/`, `lazycodex-feature-flag.ts`, `omo-codex` package, Codex plugin packaging, hook/MCP/session changes): Introduced the full Codex platform as a first-class install target alongside \"opencode\", with lazycodex marketplace routing, Codex-specific TOML config handling, bundled agent roles, lazy LSP MCP backend (on-demand start + idle timeout), start-work continuation via Codex session hooks, and `openai-codex` / `gpt-*.codex` model profiles as the deep-specialist execution lane. Telemetry, product identity (`omo-codex`), and installer shims were added (these are explicitly out-of-scope for LFG per LFP reverse-eng docs).\n27→- LFP lfe-core response (the \"check lfp lfe-core\" reference): Extracted the *neutral* contracts (AgentWorkState, IUlwHost, LazyCapability*, KeywordDetection, SharedSkill*, ReviewSynthesis/DecisionKernel, continuation directives, etc.) into `Lfe.AgentOs/`, `Lfe.Hooks/`, `Lfe.CodexAdapter/` (reference host only), `Lfe.CodexMcpBridge/`. LFP consumes them via ProjectReference + local NuGet; the pi-ai layer registered \"openai-codex\" + \"openai-codex-responses\" as a first-class KnownProvider. Duplication of OMO logic was deliberately avoided — LFE-core is the unification layer.\n28→- Current LFG \"omo-\" hierarchy (global search, excluding pure vendor/omo-standalone internals):\n29→  - `catalog/omo-skill-map.json`, references in doctor/mcp/tools/common.\n30→  - `src/runtime-ts/adapters/omo-kernel.ts` + `omo-hooks.ts` (the current OMO hook bridge).\n31→  - `omo-policy` string in supervision broker (spawn-adapter.ts).\n32→  - `omo-agent-registry-contract.json` test fixture.\n33→  - Many path references to `vendor/omo-standalone`.\n34→  - \"omo-\" prefixed tmux session helpers in dist (runtime).\n35→- Duplication risk identified: The current omo-kernel/hooks adapters + skill-map + policy strings are exactly the kind of \"omo- prefixed\" surface that LFE-core now unifies (post-4471). The plan's unification strategy is to make the new `lfe-core-*` adapters the primary implementation, with the existing omo-* files becoming thin legacy bridges that delegate to LFE contracts where possible, then delete the duplication over phases.\n36→- No other \"omo-\" product identity or installer code exists in LFG source (good — LFG correctly stayed out of OMO product surfaces).\n37→- Conclusion for unification: LFE-core (vendored) + LFG adaptation layer will be the single stack for the Agent OS features listed by the user. omo-standalone vendor stays only for model-core / ulw-kernel pieces not yet in LFE-core scope. This creates the post-4471 parity adaptation without a second hierarchy.\n38→\n39→(The planner performed the full PR-related file reads + LFP doc reads + \"omo-\" greps in both omo/lfp/lfg before writing the above. Implementer must repeat and append new findings to this section before Phase 0 code changes.)\n40→\n41→The original query (\"make things like what was done on lazycodex on the omo at ~/workspace/linalab/omo check lfp lfe-core for chekcing then do implmentatuon of that on grok with this lfg then let lfg and grok be use openai-codex as a model with checking grok build manual of adding new provider\") + the line-166 feedback = **build a full OMO-like (LFE-core-powered) harnessing Agent OS that runs *inside* Grok Build via the LFG plugin**, with:\n42→- Main orchestrator (Sisyphus + family) defaulting to **gpt-5.5 / openai-codex** (not grok-build) while Grok remains the mandatory Oracle gate.\n43→- The *exact same* adaptation layer + hook + feature architecture as OMO and LFE-core: lazy LSP (on-demand MCP), shared skills, keyword triggering logic, continuation, review synthesis, ULW host contract, etc.\n44→- **lfe-core must be used as the implementation core**, placed as a submodule / vendored tree inside the LFG package exactly like LFP structures its `lfe-core/` (local ProjectReference equivalent + contract packages).\n45→\n46→**Why the first narrow plan was insufficient:** It only solved the prerequisite model registration. The real work is the full harnessing layer (adaptation + contracts + feature parity) so that Grok Build becomes a first-class host for the LFE-core Agent OS, just as pi-agent and LazyCodex/Codex are.\n47→\n48→**Current LFG state vs. target:**\n49→- Partial OMO port (vendors omo-standalone, has runtime-ts/hooks-ts/MCP-TS, codex lane in model resolution, some smoke parity).\n50→- Missing: lfe-core contract submodule, full adaptation layer, lazy-LSP-style on-demand backends, keyword triggering, complete shared-skill + hook parity, and production use of the gpt-5.5 orchestrator lane.\n51→\n52→LFP proved the pattern works: neutral contracts in `lfe-core/`, thin host-specific adapters in `Lfp.Integration.*`, main logic consumes the contracts. LFG will do the TS/Bun equivalent on Grok Build.\n53→\n54→---\n55→\n56→## Scope (Explicitly In / Out)\n57→\n58→**In scope (phased delivery):**\n59→- Phase 0: `openai-codex` as first-class model/provider (Grok manual compliant) so the orchestrator *can* use gpt-5.5.\n60→- Phase 1: Add `lfe-core` (neutral contracts for ULW host, hooks, Agent OS, lazy capability, review, keyword, shared skills, etc.) as a vendored submodule/tree under `plugins/lfg/` (modeled exactly on LFP's layout + reference strategy).\n61→- Phase 2: Grok-specific adaptation layer (new adapters/ that implement the LFE contracts) + switch main orchestrator profiles to the openai-codex/gpt-5.5 lane (Oracle still mandatory).\n62→- Phase 3: Feature parity (lazy LSP on-demand, shared skills, keyword triggering, full hook surface) exercised through the LFE path.\n63→- All docs, evidence, AGENTS contracts, smoke/self-test updated so the claim \"Grok Build now hosts a full LFE-core OMO-like Agent OS via LFG (LFP structure)\" is true and auditable.\n64→\n65→**Out of scope:**\n66→- Re-introducing any Python LFG code or legacy grok-harnessing.\n67→- Copying OMO product surfaces (codex installer, marketplace, telemetry).\n68→- Making LFG execute C# lfe-core at runtime (unless user later chooses sidecar); v1 uses TS projections of the contracts (LFG stays pure Bun/TS).\n69→- Replacing Grok's native TUI — LFG extends it.\n70→\n71→---\n72→\n73→## Recommended Approach (LFP Structure on Grok)\n74→\n75→**Target architecture (LFE-core as unification layer — remove OMO duplication per @plan.md:14 feedback):**\n76→LFE-core (the neutral contracts extracted post-OMO PR #4471, as proven in LFP) becomes the single source of truth for the Agent OS (ULW host, hooks, lazy capability, keyword detection, shared skills, review synthesis, continuation, AgentWorkState, etc.).\n77→\n78→LFG's job is the thin Grok-specific adaptation layer on top (exactly as LFP's `Lfp.Integration.*` sits on `lfe-core/`).\n79→\n80→**Explicit goal:** Audit and *remove* \"omo-\" prefixed duplication in the code hierarchy wherever LFE-core now provides the unifying contract. The omo-standalone vendor remains only for the parts not yet unified under LFE-core (or as a temporary bridge during migration). No new parallel \"omo-\" stacks are created.\n81→\n82→**Target layout (LFP pattern + unification):**\n83→```\n84→LFG (Grok plugin)\n85→├── vendor/lfe-core/                 # THE unifying core (contracts from LFP lfe-core, post PR 4471 learnings)\n86→├── vendor-links/lfe-core/           # re-exports\n87→├── src/runtime-ts/\n88→│   ├── adapters/\n89→│   │   └── lfe-core-*.ts            # The *only* new adaptation layer (Grok-specific impl of LFE contracts)\n90→│   │   └── omo-*.ts (legacy bridge only; slated for removal as LFE unification progresses)\n91→│   └── services/ (model-resolution defaults orchestrator to openai-codex/gpt-5.5 via LFE path)\n92→└── docs/LFE-CORE-UNIFICATION.md (the migration/unification plan + duplication audit results)\n93→```\n94→\n95→**Phasing (unification first, remove duplication, land value safely):**\n96→1. **Phase 0 (small, unblocks everything):** openai-codex provider/model (Grok manual compliant). This is the prerequisite so the main orchestrator *can* use gpt-5.5.\n97→2. **Phase 1 (Unification foundation):** Add lfe-core as the submodule/unifying core (per LFP layout). Full duplication audit of all \"omo-\" code in LFG + vendor. Document in LFE-CORE-UNIFICATION.md exactly which OMO surfaces will be removed in favor of LFE-core contracts. Update ADAPTER-BOUND-DISPOSITION.md.\n98→3. **Phase 2:** Thin Grok adaptation layer on *LFE-core* (not a second parallel stack) + switch orchestrator profiles to openai-codex/gpt-5.5. Existing pure-TS paths become the \"legacy bridge\" and are marked for later removal.\n99→4. **Phase 3:** Implement the LFE-core-powered features (lazy LSP, shared skills, keyword triggering, full hooks, continuation) as the *only* path. Begin removing duplicated \"omo-\" logic.\n100→5. **Phase 4:** Full evidence on Grok Build TUI + release gates. The hierarchy is now LFE-core (core) → LFG adaptation (Grok host) with no OMO duplication for the unified surfaces.\n101→\n102→**Reuse (heavy):**\n103→- Existing vendor/ + vendor-links/ pattern for omo-standalone (copy for lfe-core).\n104→- Current runtime-ts, hooks-ts, MCP-TS, spawn-adapter, model-resolution, openai-compatible-dispatch.\n105→- Grok custom model mechanism (exactly as documented in `~/.grok/docs/user-guide/11-custom-models.md`).\n106→- LFP docs as spec (lfe-core-integration.md, lazycodex-reverse-engineering.md, architecture.md) — read-only.\n107→\n108→**Verification:** Every phase ends with full `bun test` matrix + `self-test.ts` + at least one manual Grok TUI trace exercising the new surface. Grok Oracle gate never bypassed.\n109→\n110→---\n111→\n112→## Critical Files (High-Level)\n113→\n114→**Edit (existing LFG):**\n115→- `plugins/lfg/src/runtime-ts/services/model-resolution.ts` + tests (Phase 0 + 2)\n116→- `plugins/lfg/src/runtime-ts/commands/provider.*` + common.ts\n117→- `plugins/lfg/config/grok-config-template.toml` (add [model.openai-codex] with responses)\n118→- `plugins/lfg/src/runtime-ts/services/openai-compatible-dispatch.ts` + spawn-adapter.ts\n119→- `plugins/lfg/docs/ARCHITECTURE.md`, `SMOKE.md`, `skills/provider/SKILL.md`\n120→- `plugins/lfg/bin/self-test.ts`, smoke-ts checks, fixtures\n121→- All 5 `AGENTS.md` (structural updates only)\n122→- `plugins/lfg/src/runtime-ts/adapters/*` (add lfe-core siblings)\n123→\n124→**New:**\n125→- `plugins/lfg/vendor/lfe-core/` (contracts)\n126→- `plugins/lfg/vendor-links/lfe-core/`\n127→- `plugins/lfg/src/runtime-ts/adapters/lfe-core-*.ts` (several)\n128→- `plugins/lfg/docs/LFE-CORE-INTEGRATION.md`\n129→- `plugins/lfg/docs/lazycodex-lfe-parity-audit.md` (the \"check lfp lfe-core\" evidence)\n130→\n131→**Read-only during work (and the implementer MUST re-audit these before every phase):**\n132→- OMO PR #4471 in full (all files, tests, docs, omo-codex package) — this is the primary source of \"what was done on lazycodex\".\n133→- Everything in `~/workspace/linalab/lfp/lfe-core/` and its docs (lazycodex-reverse-engineering.md, lfe-core-integration.md, lfe-core-agent-os-consumption.md, the actual Lfe.CodexAdapter / Lfe.CodexMcpBridge / Lfe.AgentOs sources) — the \"check lfp lfe-core for checking\" reference.\n134→- `~/.grok/docs/user-guide/11-custom-models.md` (exact TOML + api_backend rules for openai-codex).\n135→- Current LFG vendor/omo-standalone + runtime-ts + hooks-ts (to find the exact insertion points for the LFE adaptation layer).\n136→\n137→The plan author re-audited the above (via list_dir, multiple greps, and targeted reads of the key PR-related paths and LFP docs) before finalizing this version. The implementer must repeat the audit.\n138→\n139→---\n140→\n141→## Verification (Phased, Strict)\n142→\n143→**Phase 0 gate (openai-codex model):** `lfg provider add --kind openai-codex`, spawn with that provider, Grok config.toml section per manual, all tests + self-test green, docs updated.\n144→\n145→**Phase 1 gate:** lfe-core tree present and importable; no regression in OMO vendor; integration doc written.\n146→\n147→**Phase 2+ gate:** End-to-end manual run on Grok TUI where Sisyphus/Hephaestus etc. resolve to gpt-5.5 via LFE-core adaptation layer, lazy-LSP / keyword / shared-skill features fire, evidence recorded, Oracle still required. Full test matrix + self-test pass. All AGENTS.md complete.\n148→\n149→**Global:** Only files under `plugins/lfg/` change. No Python, no weakening of existing contracts or Oracle gate.\n150→"}}],"rawOutput":{"type":"ReadFile","FileContent":{"content":"1→# Implementation Plan: OMO-like Harnessing Agent OS on Grok Build (LFE-core + LFG Adaptation Layer + GPT-5.5 Orchestrator)\n2→\n3→**Session:** 019e7466-d18d-71b0-9b1f-6599dc5defbc  \n4→**Date:** 2026-05-30 (revised after user clarification)  \n5→**Status:** Ready for user approval (Plan Mode) — **Broad scope: full LFE-core-powered OMO harnessing Agent OS on Grok Build**\n6→\n7→---\n8→\n9→## Context (User's Clarified Vision — \"in short\")\n10→\n11→**User's direct clarification:** \"in short is this the plan to creat omo like harnessing agent os that runs on grok build with supporting main ochestrator be using gpt5.5 not grok build on grok build with same manner of adaptation layer and hook and featurew like lazy lsp and shared skills and keyword triggering logic with omo and lfe-core must use lfe-core to implment this feature put lfe-core as submodules like lfp\"\n12→\n13→**Yes — this revised plan is exactly that.**\n14→\n15→**MANDATORY FIRST STEP FOR ANY IMPLEMENTER (per user feedback at plan review — including @plan.md:14):** Before writing *any* code, the person executing this plan **must**:\n16→1. `gh pr view 4471 --repo code-yeongyu/oh-my-openagent --json files,commits,body` (or equivalent full diff + discussion) + read *every* related doc and \"omo-\" file touched by the PR.\n17→2. Globally search the LFG codebase (and the omo-standalone vendor) for any \"omo-\" prefixed duplication in the code hierarchy (agent definitions, hook shapes, skill catalogs, LSP/MCP lazy logic, shared skills, keyword triggering, continuation, review synthesis, etc.).\n18→3. Re-read LFP's `docs/lazycodex-reverse-engineering.md`, `lfe-core-integration.md`, `lfe-core-agent-os-consumption.md`, and the actual `lfe-core/src/Lfe.CodexAdapter/` + `Lfe.CodexMcpBridge/` + `Lfe.AgentOs/` implementations.\n19→4. **Decide and document** how LFE-core becomes the single unifying core (removing OMO duplication where LFE-core now provides the neutral contract). The adaptation layer created after PR #4471 learnings must sit on top of LFE-core, not duplicate OMO logic.\n20→5. Re-audit *this plan.md* against the above and update it with the unification strategy before any implementation starts.\n21→6. Only then begin Phase 0.\n22→\n23→This \"check OMO PR 4471 + remove omo- duplication + use lfe-core for unification + check your plan again\" gate is non-negotiable and must be evidenced in the first commit or PR description. LFE-core is the unification path to OMO parity (post-4471), not a second parallel stack.\n24→\n25→**Initial Audit Findings (executed by planner + re-checked per user @plan.md:14 feedback, 2026-05-30):**\n26→- OMO PR 4471 artifacts (from exploration of the changed files in `src/cli/install-codex/`, `lazycodex-feature-flag.ts`, `omo-codex` package, Codex plugin packaging, hook/MCP/session changes): Introduced the full Codex platform as a first-class install target alongside \"opencode\", with lazycodex marketplace routing, Codex-specific TOML config handling, bundled agent roles, lazy LSP MCP backend (on-demand start + idle timeout), start-work continuation via Codex session hooks, and `openai-codex` / `gpt-*.codex` model profiles as the deep-specialist execution lane. Telemetry, product identity (`omo-codex`), and installer shims were added (these are explicitly out-of-scope for LFG per LFP reverse-eng docs).\n27→- LFP lfe-core response (the \"check lfp lfe-core\" reference): Extracted the *neutral* contracts (AgentWorkState, IUlwHost, LazyCapability*, KeywordDetection, SharedSkill*, ReviewSynthesis/DecisionKernel, continuation directives, etc.) into `Lfe.AgentOs/`, `Lfe.Hooks/`, `Lfe.CodexAdapter/` (reference host only), `Lfe.CodexMcpBridge/`. LFP consumes them via ProjectReference + local NuGet; the pi-ai layer registered \"openai-codex\" + \"openai-codex-responses\" as a first-class KnownProvider. Duplication of OMO logic was deliberately avoided — LFE-core is the unification layer.\n28→- Current LFG \"omo-\" hierarchy (global search, excluding pure vendor/omo-standalone internals):\n29→  - `catalog/omo-skill-map.json`, references in doctor/mcp/tools/common.\n30→  - `src/runtime-ts/adapters/omo-kernel.ts` + `omo-hooks.ts` (the current OMO hook bridge).\n31→  - `omo-policy` string in supervision broker (spawn-adapter.ts).\n32→  - `omo-agent-registry-contract.json` test fixture.\n33→  - Many path references to `vendor/omo-standalone`.\n34→  - \"omo-\" prefixed tmux session helpers in dist (runtime).\n35→- Duplication risk identified: The current omo-kernel/hooks adapters + skill-map + policy strings are exactly the kind of \"omo- prefixed\" surface that LFE-core now unifies (post-4471). The plan's unification strategy is to make the new `lfe-core-*` adapters the primary implementation, with the existing omo-* files becoming thin legacy bridges that delegate to LFE contracts where possible, then delete the duplication over phases.\n36→- No other \"omo-\" product identity or installer code exists in LFG source (good — LFG correctly stayed out of OMO product surfaces).\n37→- Conclusion for unification: LFE-core (vendored) + LFG adaptation layer will be the single stack for the Agent OS features listed by the user. omo-standalone vendor stays only for model-core / ulw-kernel pieces not yet in LFE-core scope. This creates the post-4471 parity adaptation without a second hierarchy.\n38→\n39→(The planner performed the full PR-related file reads + LFP doc reads + \"omo-\" greps in both omo/lfp/lfg before writing the above. Implementer must repeat and append new findings to this section before Phase 0 code changes.)\n40→\n41→The original query (\"make things like what was done on lazycodex on the omo at ~/workspace/linalab/omo check lfp lfe-core for chekcing then do implmentatuon of that on grok with this lfg then let lfg and grok be use openai-codex as a model with checking grok build manual of adding new provider\") + the line-166 feedback = **build a full OMO-like (LFE-core-powered) harnessing Agent OS that runs *inside* Grok Build via the LFG plugin**, with:\n42→- Main orchestrator (Sisyphus + family) defaulting to **gpt-5.5 / openai-codex** (not grok-build) while Grok remains the mandatory Oracle gate.\n43→- The *exact same* adaptation layer + hook + feature architecture as OMO and LFE-core: lazy LSP (on-demand MCP), shared skills, keyword triggering logic, continuation, review synthesis, ULW host contract, etc.\n44→- **lfe-core must be used as the implementation core**, placed as a submodule / vendored tree inside the LFG package exactly like LFP structures its `lfe-core/` (local ProjectReference equivalent + contract packages).\n45→\n46→**Why the first narrow plan was insufficient:** It only solved the prerequisite model registration. The real work is the full harnessing layer (adaptation + contracts + feature parity) so that Grok Build becomes a first-class host for the LFE-core Agent OS, just as pi-agent and LazyCodex/Codex are.\n47→\n48→**Current LFG state vs. target:**\n49→- Partial OMO port (vendors omo-standalone, has runtime-ts/hooks-ts/MCP-TS, codex lane in model resolution, some smoke parity).\n50→- Missing: lfe-core contract submodule, full adaptation layer, lazy-LSP-style on-demand backends, keyword triggering, complete shared-skill + hook parity, and production use of the gpt-5.5 orchestrator lane.\n51→\n52→LFP proved the pattern works: neutral contracts in `lfe-core/`, thin host-specific adapters in `Lfp.Integration.*`, main logic consumes the contracts. LFG will do the TS/Bun equivalent on Grok Build.\n53→\n54→---\n55→\n56→## Scope (Explicitly In / Out)\n57→\n58→**In scope (phased delivery):**\n59→- Phase 0: `openai-codex` as first-class model/provider (Grok manual compliant) so the orchestrator *can* use gpt-5.5.\n60→- Phase 1: Add `lfe-core` (neutral contracts for ULW host, hooks, Agent OS, lazy capability, review, keyword, shared skills, etc.) as a vendored submodule/tree under `plugins/lfg/` (modeled exactly on LFP's layout + reference strategy).\n61→- Phase 2: Grok-specific adaptation layer (new adapters/ that implement the LFE contracts) + switch main orchestrator profiles to the openai-codex/gpt-5.5 lane (Oracle still mandatory).\n62→- Phase 3: Feature parity (lazy LSP on-demand, shared skills, keyword triggering, full hook surface) exercised through the LFE path.\n63→- All docs, evidence, AGENTS contracts, smoke/self-test updated so the claim \"Grok Build now hosts a full LFE-core OMO-like Agent OS via LFG (LFP structure)\" is true and auditable.\n64→\n65→**Out of scope:**\n66→- Re-introducing any Python LFG code or legacy grok-harnessing.\n67→- Copying OMO product surfaces (codex installer, marketplace, telemetry).\n68→- Making LFG execute C# lfe-core at runtime (unless user later chooses sidecar); v1 uses TS projections of the contracts (LFG stays pure Bun/TS).\n69→- Replacing Grok's native TUI — LFG extends it.\n70→\n71→---\n72→\n73→## Recommended Approach (LFP Structure on Grok)\n74→\n75→**Target architecture (LFE-core as unification layer — remove OMO duplication per @plan.md:14 feedback):**\n76→LFE-core (the neutral contracts extracted post-OMO PR #4471, as proven in LFP) becomes the single source of truth for the Agent OS (ULW host, hooks, lazy capability, keyword detection, shared skills, review synthesis, continuation, AgentWorkState, etc.).\n77→\n78→LFG's job is the thin Grok-specific adaptation layer on top (exactly as LFP's `Lfp.Integration.*` sits on `lfe-core/`).\n79→\n80→**Explicit goal:** Audit and *remove* \"omo-\" prefixed duplication in the code hierarchy wherever LFE-core now provides the unifying contract. The omo-standalone vendor remains only for the parts not yet unified under LFE-core (or as a temporary bridge during migration). No new parallel \"omo-\" stacks are created.\n81→\n82→**Target layout (LFP pattern + unification):**\n83→```\n84→LFG (Grok plugin)\n85→├── vendor/lfe-core/                 # THE unifying core (contracts from LFP lfe-core, post PR 4471 learnings)\n86→├── vendor-links/lfe-core/           # re-exports\n87→├── src/runtime-ts/\n88→│   ├── adapters/\n89→│   │   └── lfe-core-*.ts            # The *only* new adaptation layer (Grok-specific impl of LFE contracts)\n90→│   │   └── omo-*.ts (legacy bridge only; slated for removal as LFE unification progresses)\n91→│   └── services/ (model-resolution defaults orchestrator to openai-codex/gpt-5.5 via LFE path)\n92→└── docs/LFE-CORE-UNIFICATION.md (the migration/unification plan + duplication audit results)\n93→```\n94→\n95→**Phasing (unification first, remove duplication, land value safely):**\n96→1. **Phase 0 (small, unblocks everything):** openai-codex provider/model (Grok manual compliant). This is the prerequisite so the main orchestrator *can* use gpt-5.5.\n97→2. **Phase 1 (Unification foundation):** Add lfe-core as the submodule/unifying core (per LFP layout). Full duplication audit of all \"omo-\" code in LFG + vendor. Document in LFE-CORE-UNIFICATION.md exactly which OMO surfaces will be removed in favor of LFE-core contracts. Update ADAPTER-BOUND-DISPOSITION.md.\n98→3. **Phase 2:** Thin Grok adaptation layer on *LFE-core* (not a second parallel stack) + switch orchestrator profiles to openai-codex/gpt-5.5. Existing pure-TS paths become the \"legacy bridge\" and are marked for later removal.\n99→4. **Phase 3:** Implement the LFE-core-powered features (lazy LSP, shared skills, keyword triggering, full hooks, continuation) as the *only* path. Begin removing duplicated \"omo-\" logic.\n100→5. **Phase 4:** Full evidence on Grok Build TUI + release gates. The hierarchy is now LFE-core (core) → LFG adaptation (Grok host) with no OMO duplication for the unified surfaces.\n101→\n102→**Reuse (heavy):**\n103→- Existing vendor/ + vendor-links/ pattern for omo-standalone (copy for lfe-core).\n104→- Current runtime-ts, hooks-ts, MCP-TS, spawn-adapter, model-resolution, openai-compatible-dispatch.\n105→- Grok custom model mechanism (exactly as documented in `~/.grok/docs/user-guide/11-custom-models.md`).\n106→- LFP docs as spec (lfe-core-integration.md, lazycodex-reverse-engineering.md, architecture.md) — read-only.\n107→\n108→**Verification:** Every phase ends with full `bun test` matrix + `self-test.ts` + at least one manual Grok TUI trace exercising the new surface. Grok Oracle gate never bypassed.\n109→\n110→---\n111→\n112→## Critical Files (High-Level)\n113→\n114→**Edit (existing LFG):**\n115→- `plugins/lfg/src/runtime-ts/services/model-resolution.ts` + tests (Phase 0 + 2)\n116→- `plugins/lfg/src/runtime-ts/commands/provider.*` + common.ts\n117→- `plugins/lfg/config/grok-config-template.toml` (add [model.openai-codex] with responses)\n118→- `plugins/lfg/src/runtime-ts/services/openai-compatible-dispatch.ts` + spawn-adapter.ts\n119→- `plugins/lfg/docs/ARCHITECTURE.md`, `SMOKE.md`, `skills/provider/SKILL.md`\n120→- `plugins/lfg/bin/self-test.ts`, smoke-ts checks, fixtures\n121→- All 5 `AGENTS.md` (structural updates only)\n122→- `plugins/lfg/src/runtime-ts/adapters/*` (add lfe-core siblings)\n123→\n124→**New:**\n125→- `plugins/lfg/vendor/lfe-core/` (contracts)\n126→- `plugins/lfg/vendor-links/lfe-core/`\n127→- `plugins/lfg/src/runtime-ts/adapters/lfe-core-*.ts` (several)\n128→- `plugins/lfg/docs/LFE-CORE-INTEGRATION.md`\n129→- `plugins/lfg/docs/lazycodex-lfe-parity-audit.md` (the \"check lfp lfe-core\" evidence)\n130→\n131→**Read-only during work (and the implementer MUST re-audit these before every phase):**\n132→- OMO PR #4471 in full (all files, tests, docs, omo-codex package) — this is the primary source of \"what was done on lazycodex\".\n133→- Everything in `~/workspace/linalab/lfp/lfe-core/` and its docs (lazycodex-reverse-engineering.md, lfe-core-integration.md, lfe-core-agent-os-consumption.md, the actual Lfe.CodexAdapter / Lfe.CodexMcpBridge / Lfe.AgentOs sources) — the \"check lfp lfe-core for checking\" reference.\n134→- `~/.grok/docs/user-guide/11-custom-models.md` (exact TOML + api_backend rules for openai-codex).\n135→- Current LFG vendor/omo-standalone + runtime-ts + hooks-ts (to find the exact insertion points for the LFE adaptation layer).\n136→\n137→The plan author re-audited the above (via list_dir, multiple greps, and targeted reads of the key PR-related paths and LFP docs) before finalizing this version. The implementer must repeat the audit.\n138→\n139→---\n140→\n141→## Verification (Phased, Strict)\n142→\n143→**Phase 0 gate (openai-codex model):** `lfg provider add --kind openai-codex`, spawn with that provider, Grok config.toml section per manual, all tests + self-test green, docs updated.\n144→\n145→**Phase 1 gate:** lfe-core tree present and importable; no regression in OMO vendor; integration doc written.\n146→\n147→**Phase 2+ gate:** End-to-end manual run on Grok TUI where Sisyphus/Hephaestus etc. resolve to gpt-5.5 via LFE-core adaptation layer, lazy-LSP / keyword / shared-skill features fire, evidence recorded, Oracle still required. Full test matrix + self-test pass. All AGENTS.md complete.\n148→\n149→**Global:** Only files under `plugins/lfg/` change. No Python, no weakening of existing contracts or Oracle gate.\n150→","content_concise":"1→# Implementation Plan: OMO-like Harnessing Agent OS on Grok Build (LFE-core + LFG Adaptation Layer + GPT-5.5 Orchestrator)\n2→\n3→**Session:** 019e7466-d18d-71b0-9b1f-6599dc5defbc  \n4→**Date:** 2026-05-30 (revised after user clarification)  \n5→**Status:** Ready for user approval (Plan Mode) — **Broad scope: full LFE-core-powered OMO harnessing Agent OS on Grok Build**\n6→\n7→---\n8→\n9→## Context (User's Clarified Vision — \"in short\")\n10→\n11→**User's direct clarification:** \"in short is this the plan to creat omo like harnessing agent os that runs on grok build with supporting main ochestrator be using gpt5.5 not grok build on grok build with same manner of adaptation layer and hook and featurew like lazy lsp and shared skills and keyword triggering logic with omo and lfe-core must use lfe-core to implment this feature put lfe-core as submodules like lfp\"\n12→\n13→**Yes — this revised plan is exactly that.**\n14→\n15→**MANDATORY FIRST STEP FOR ANY IMPLEMENTER (per user feedback at plan review — including @plan.md:14):** Before writing *any* code, the person executing this plan **must**:\n16→1. `gh pr view 4471 --repo code-yeongyu/oh-my-openagent --json files,commits,body` (or equivalent full diff + discussion) + read *every* related doc and \"omo-\" file touched by the PR.\n17→2. Globally search the LFG codebase (and the omo-standalone vendor) for any \"omo-\" prefixed duplication in the code hierarchy (agent definitions, hook shapes, skill catalogs, LSP/MCP lazy logic, shared skills, keyword triggering, continuation, review synthesis, etc.).\n18→3. Re-read LFP's `docs/lazycodex-reverse-engineering.md`, `lfe-core-integration.md`, `lfe-core-agent-os-consumption.md`, and the actual `lfe-core/src/Lfe.CodexAdapter/` + `Lfe.CodexMcpBridge/` + `Lfe.AgentOs/` implementations.\n19→4. **Decide and document** how LFE-core becomes the single unifying core (removing OMO duplication where LFE-core now provides the neutral contract). The adaptation layer created after PR #4471 learnings must sit on top of LFE-core, not duplicate OMO logic.\n20→5. Re-audit *this plan.md* against the above and update it with the unification strategy before any implementation starts.\n21→6. Only then begin Phase 0.\n22→\n23→This \"check OMO PR 4471 + remove omo- duplication + use lfe-core for unification + check your plan again\" gate is non-negotiable and must be evidenced in the first commit or PR description. LFE-core is the unification path to OMO parity (post-4471), not a second parallel stack.\n24→\n25→**Initial Audit Findings (executed by planner + re-checked per user @plan.md:14 feedback, 2026-05-30):**\n26→- OMO PR 4471 artifacts (from exploration of the changed files in `src/cli/install-codex/`, `lazycodex-feature-flag.ts`, `omo-codex` package, Codex plugin packaging, hook/MCP/session changes): Introduced the full Codex platform as a first-class install target alongside \"opencode\", with lazycodex marketplace routing, Codex-specific TOML config handling, bundled agent roles, lazy LSP MCP backend (on-demand start + idle timeout), start-work continuation via Codex session hooks, and `openai-codex` / `gpt-*.codex` model profiles as the deep-specialist execution lane. Telemetry, product identity (`omo-codex`), and installer shims were added (these are explicitly out-of-scope for LFG per LFP reverse-eng docs).\n27→- LFP lfe-core response (the \"check lfp lfe-core\" reference): Extracted the *neutral* contracts (AgentWorkState, IUlwHost, LazyCapability*, KeywordDetection, SharedSkill*, ReviewSynthesis/DecisionKernel, continuation directives, etc.) into `Lfe.AgentOs/`, `Lfe.Hooks/`, `Lfe.CodexAdapter/` (reference host only), `Lfe.CodexMcpBridge/`. LFP consumes them via ProjectReference + local NuGet; the pi-ai layer registered \"openai-codex\" + \"openai-codex-responses\" as a first-class KnownProvider. Duplication of OMO logic was deliberately avoided — LFE-core is the unification layer.\n28→- Current LFG \"omo-\" hierarchy (global search, excluding pure vendor/omo-standalone internals):\n29→  - `catalog/omo-skill-map.json`, references in doctor/mcp/tools/common.\n30→  - `src/runtime-ts/adapters/omo-kernel.ts` + `omo-hooks.ts` (the current OMO hook bridge).\n31→  - `omo-policy` string in supervision broker (spawn-adapter.ts).\n32→  - `omo-agent-registry-contract.json` test fixture.\n33→  - Many path references to `vendor/omo-standalone`.\n34→  - \"omo-\" prefixed tmux session helpers in dist (runtime).\n35→- Duplication risk identified: The current omo-kernel/hooks adapters + skill-map + policy strings are exactly the kind of \"omo- prefixed\" surface that LFE-core now unifies (post-4471). The plan's unification strategy is to make the new `lfe-core-*` adapters the primary implementation, with the existing omo-* files becoming thin legacy bridges that delegate to LFE contracts where possible, then delete the duplication over phases.\n36→- No other \"omo-\" product identity or installer code exists in LFG source (good — LFG correctly stayed out of OMO product surfaces).\n37→- Conclusion for unification: LFE-core (vendored) + LFG adaptation layer will be the single stack for the Agent OS features listed by the user. omo-standalone vendor stays only for model-core / ulw-kernel pieces not yet in LFE-core scope. This creates the post-4471 parity adaptation without a second hierarchy.\n38→\n39→(The planner performed the full PR-related file reads + LFP doc reads + \"omo-\" greps in both omo/lfp/lfg before writing the above. Implementer must repeat and append new findings to this section before Phase 0 code changes.)\n40→\n41→The original query (\"make things like what was done on lazycodex on the omo at ~/workspace/linalab/omo check lfp lfe-core for chekcing then do implmentatuon of that on grok with this lfg then let lfg and grok be use openai-codex as a model with checking grok build manual of adding new provider\") + the line-166 feedback = **build a full OMO-like (LFE-core-powered) harnessing Agent OS that runs *inside* Grok Build via the LFG plugin**, with:\n42→- Main orchestrator (Sisyphus + family) defaulting to **gpt-5.5 / openai-codex** (not grok-build) while Grok remains the mandatory Oracle gate.\n43→- The *exact same* adaptation layer + hook + feature architecture as OMO and LFE-core: lazy LSP (on-demand MCP), shared skills, keyword triggering logic, continuation, review synthesis, ULW host contract, etc.\n44→- **lfe-core must be used as the implementation core**, placed as a submodule / vendored tree inside the LFG package exactly like LFP structures its `lfe-core/` (local ProjectReference equivalent + contract packages).\n45→\n46→**Why the first narrow plan was insufficient:** It only solved the prerequisite model registration. The real work is the full harnessing layer (adaptation + contracts + feature parity) so that Grok Build becomes a first-class host for the LFE-core Agent OS, just as pi-agent and LazyCodex/Codex are.\n47→\n48→**Current LFG state vs. target:**\n49→- Partial OMO port (vendors omo-standalone, has runtime-ts/hooks-ts/MCP-TS, codex lane in model resolution, some smoke parity).\n50→- Missing: lfe-core contract submodule, full adaptation layer, lazy-LSP-style on-demand backends, keyword triggering, complete shared-skill + hook parity, and production use of the gpt-5.5 orchestrator lane.\n51→\n52→LFP proved the pattern works: neutral contracts in `lfe-core/`, thin host-specific adapters in `Lfp.Integration.*`, main logic consumes the contracts. LFG will do the TS/Bun equivalent on Grok Build.\n53→\n54→---\n55→\n56→## Scope (Explicitly In / Out)\n57→\n58→**In scope (phased delivery):**\n59→- Phase 0: `openai-codex` as first-class model/provider (Grok manual compliant) so the orchestrator *can* use gpt-5.5.\n60→- Phase 1: Add `lfe-core` (neutral contracts for ULW host, hooks, Agent OS, lazy capability, review, keyword, shared skills, etc.) as a vendored submodule/tree under `plugins/lfg/` (modeled exactly on LFP's layout + reference strategy).\n61→- Phase 2: Grok-specific adaptation layer (new adapters/ that implement the LFE contracts) + switch main orchestrator profiles to the openai-codex/gpt-5.5 lane (Oracle still mandatory).\n62→- Phase 3: Feature parity (lazy LSP on-demand, shared skills, keyword triggering, full hook surface) exercised through the LFE path.\n63→- All docs, evidence, AGENTS contracts, smoke/self-test updated so the claim \"Grok Build now hosts a full LFE-core OMO-like Agent OS via LFG (LFP structure)\" is true and auditable.\n64→\n65→**Out of scope:**\n66→- Re-introducing any Python LFG code or legacy grok-harnessing.\n67→- Copying OMO product surfaces (codex installer, marketplace, telemetry).\n68→- Making LFG execute C# lfe-core at runtime (unless user later chooses sidecar); v1 uses TS projections of the contracts (LFG stays pure Bun/TS).\n69→- Replacing Grok's native TUI — LFG extends it.\n70→\n71→---\n72→\n73→## Recommended Approach (LFP Structure on Grok)\n74→\n75→**Target architecture (LFE-core as unification layer — remove OMO duplication per @plan.md:14 feedback):**\n76→LFE-core (the neutral contracts extracted post-OMO PR #4471, as proven in LFP) becomes the single source of truth for the Agent OS (ULW host, hooks, lazy capability, keyword detection, shared skills, review synthesis, continuation, AgentWorkState, etc.).\n77→\n78→LFG's job is the thin Grok-specific adaptation layer on top (exactly as LFP's `Lfp.Integration.*` sits on `lfe-core/`).\n79→\n80→**Explicit goal:** Audit and *remove* \"omo-\" prefixed duplication in the code hierarchy wherever LFE-core now provides the unifying contract. The omo-standalone vendor remains only for the parts not yet unified under LFE-core (or as a temporary bridge during migration). No new parallel \"omo-\" stacks are created.\n81→\n82→**Target layout (LFP pattern + unification):**\n83→```\n84→LFG (Grok plugin)\n85→├── vendor/lfe-core/                 # THE unifying core (contracts from LFP lfe-core, post PR 4471 learnings)\n86→├── vendor-links/lfe-core/           # re-exports\n87→├── src/runtime-ts/\n88→│   ├── adapters/\n89→│   │   └── lfe-core-*.ts            # The *only* new adaptation layer (Grok-specific impl of LFE contracts)\n90→│   │   └── omo-*.ts (legacy bridge only; slated for removal as LFE unification progresses)\n91→│   └── services/ (model-resolution defaults orchestrator to openai-codex/gpt-5.5 via LFE path)\n92→└── docs/LFE-CORE-UNIFICATION.md (the migration/unification plan + duplication audit results)\n93→```\n94→\n95→**Phasing (unification first, remove duplication, land value safely):**\n96→1. **Phase 0 (small, unblocks everything):** openai-codex provider/model (Grok manual compliant). This is the prerequisite so the main orchestrator *can* use gpt-5.5.\n97→2. **Phase 1 (Unification foundation):** Add lfe-core as the submodule/unifying core (per LFP layout). Full duplication audit of all \"omo-\" code in LFG + vendor. Document in LFE-CORE-UNIFICATION.md exactly which OMO surfaces will be removed in favor of LFE-core contracts. Update ADAPTER-BOUND-DISPOSITION.md.\n98→3. **Phase 2:** Thin Grok adaptation layer on *LFE-core* (not a second parallel stack) + switch orchestrator profiles to openai-codex/gpt-5.5. Existing pure-TS paths become the \"legacy bridge\" and are marked for later removal.\n99→4. **Phase 3:** Implement the LFE-core-powered features (lazy LSP, shared skills, keyword triggering, full hooks, continuation) as the *only* path. Begin removing duplicated \"omo-\" logic.\n100→5. **Phase 4:** Full evidence on Grok Build TUI + release gates. The hierarchy is now LFE-core (core) → LFG adaptation (Grok host) with no OMO duplication for the unified surfaces.\n101→\n102→**Reuse (heavy):**\n103→- Existing vendor/ + vendor-links/ pattern for omo-standalone (copy for lfe-core).\n104→- Current runtime-ts, hooks-ts, MCP-TS, spawn-adapter, model-resolution, openai-compatible-dispatch.\n105→- Grok custom model mechanism (exactly as documented in `~/.grok/docs/user-guide/11-custom-models.md`).\n106→- LFP docs as spec (lfe-core-integration.md, lazycodex-reverse-engineering.md, architecture.md) — read-only.\n107→\n108→**Verification:** Every phase ends with full `bun test` matrix + `self-test.ts` + at least one manual Grok TUI trace exercising the new surface. Grok Oracle gate never bypassed.\n109→\n110→---\n111→\n112→## Critical Files (High-Level)\n113→\n114→**Edit (existing LFG):**\n115→- `plugins/lfg/src/runtime-ts/services/model-resolution.ts` + tests (Phase 0 + 2)\n116→- `plugins/lfg/src/runtime-ts/commands/provider.*` + common.ts\n117→- `plugins/lfg/config/grok-config-template.toml` (add [model.openai-codex] with responses)\n118→- `plugins/lfg/src/runtime-ts/services/openai-compatible-dispatch.ts` + spawn-adapter.ts\n119→- `plugins/lfg/docs/ARCHITECTURE.md`, `SMOKE.md`, `skills/provider/SKILL.md`\n120→- `plugins/lfg/bin/self-test.ts`, smoke-ts checks, fixtures\n121→- All 5 `AGENTS.md` (structural updates only)\n122→- `plugins/lfg/src/runtime-ts/adapters/*` (add lfe-core siblings)\n123→\n124→**New:**\n125→- `plugins/lfg/vendor/lfe-core/` (contracts)\n126→- `plugins/lfg/vendor-links/lfe-core/`\n127→- `plugins/lfg/src/runtime-ts/adapters/lfe-core-*.ts` (several)\n128→- `plugins/lfg/docs/LFE-CORE-INTEGRATION.md`\n129→- `plugins/lfg/docs/lazycodex-lfe-parity-audit.md` (the \"check lfp lfe-core\" evidence)\n130→\n131→**Read-only during work (and the implementer MUST re-audit these before every phase):**\n132→- OMO PR #4471 in full (all files, tests, docs, omo-codex package) — this is the primary source of \"what was done on lazycodex\".\n133→- Everything in `~/workspace/linalab/lfp/lfe-core/` and its docs (lazycodex-reverse-engineering.md, lfe-core-integration.md, lfe-core-agent-os-consumption.md, the actual Lfe.CodexAdapter / Lfe.CodexMcpBridge / Lfe.AgentOs sources) — the \"check lfp lfe-core for checking\" reference.\n134→- `~/.grok/docs/user-guide/11-custom-models.md` (exact TOML + api_backend rules for openai-codex).\n135→- Current LFG vendor/omo-standalone + runtime-ts + hooks-ts (to find the exact insertion points for the LFE adaptation layer).\n136→\n137→The plan author re-audited the above (via list_dir, multiple greps, and targeted reads of the key PR-related paths and LFP docs) before finalizing this version. The implementer must repeat the audit.\n138→\n139→---\n140→\n141→## Verification (Phased, Strict)\n142→\n143→**Phase 0 gate (openai-codex model):** `lfg provider add --kind openai-codex`, spawn with that provider, Grok config.toml section per manual, all tests + self-test green, docs updated.\n144→\n145→**Phase 1 gate:** lfe-core tree present and importable; no regression in OMO vendor; integration doc written.\n146→\n147→**Phase 2+ gate:** End-to-end manual run on Grok TUI where Sisyphus/Hephaestus etc. resolve to gpt-5.5 via LFE-core adaptation layer, lazy-LSP / keyword / shared-skill features fire, evidence recorded, Oracle still required. Full test matrix + self-test pass. All AGENTS.md complete.\n148→\n149→**Global:** Only files under `plugins/lfg/` change. No Python, no weakening of existing contracts or Oracle gate.\n150→","absolute_path":"/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e7466-d18d-71b0-9b1f-6599dc5defbc/plan.md","offset":1,"limit":150,"raw_output":"# Implementation Plan: OMO-like Harnessing Agent OS on Grok Build (LFE-core + LFG Adaptation Layer + GPT-5.5 Orchestrator)\n\n**Session:** 019e7466-d18d-71b0-9b1f-6599dc5defbc  \n**Date:** 2026-05-30 (revised after user clarification)  \n**Status:** Ready for user approval (Plan Mode) — **Broad scope: full LFE-core-powered OMO harnessing Agent OS on Grok Build**\n\n---\n\n## Context (User's Clarified Vision — \"in short\")\n\n**User's direct clarification:** \"in short is this the plan to creat omo like harnessing agent os that runs on grok build with supporting main ochestrator be using gpt5.5 not grok build on grok build with same manner of adaptation layer and hook and featurew like lazy lsp and shared skills and keyword triggering logic with omo and lfe-core must use lfe-core to implment this feature put lfe-core as submodules like lfp\"\n\n**Yes — this revised plan is exactly that.**\n\n**MANDATORY FIRST STEP FOR ANY IMPLEMENTER (per user feedback at plan review — including @plan.md:14):** Before writing *any* code, the person executing this plan **must**:\n1. `gh pr view 4471 --repo code-yeongyu/oh-my-openagent --json files,commits,body` (or equivalent full diff + discussion) + read *every* related doc and \"omo-\" file touched by the PR.\n2. Globally search the LFG codebase (and the omo-standalone vendor) for any \"omo-\" prefixed duplication in the code hierarchy (agent definitions, hook shapes, skill catalogs, LSP/MCP lazy logic, shared skills, keyword triggering, continuation, review synthesis, etc.).\n3. Re-read LFP's `docs/lazycodex-reverse-engineering.md`, `lfe-core-integration.md`, `lfe-core-agent-os-consumption.md`, and the actual `lfe-core/src/Lfe.CodexAdapter/` + `Lfe.CodexMcpBridge/` + `Lfe.AgentOs/` implementations.\n4. **Decide and document** how LFE-core becomes the single unifying core (removing OMO duplication where LFE-core now provides the neutral contract). The adaptation layer created after PR #4471 learnings must sit on top of LFE-core, not duplicate OMO logic.\n5. Re-audit *this plan.md* against the above and update it with the unification strategy before any implementation starts.\n6. Only then begin Phase 0.\n\nThis \"check OMO PR 4471 + remove omo- duplication + use lfe-core for unification + check your plan again\" gate is non-negotiable and must be evidenced in the first commit or PR description. LFE-core is the unification path to OMO parity (post-4471), not a second parallel stack.\n\n**Initial Audit Findings (executed by planner + re-checked per user @plan.md:14 feedback, 2026-05-30):**\n- OMO PR 4471 artifacts (from exploration of the changed files in `src/cli/install-codex/`, `lazycodex-feature-flag.ts`, `omo-codex` package, Codex plugin packaging, hook/MCP/session changes): Introduced the full Codex platform as a first-class install target alongside \"opencode\", with lazycodex marketplace routing, Codex-specific TOML config handling, bundled agent roles, lazy LSP MCP backend (on-demand start + idle timeout), start-work continuation via Codex session hooks, and `openai-codex` / `gpt-*.codex` model profiles as the deep-specialist execution lane. Telemetry, product identity (`omo-codex`), and installer shims were added (these are explicitly out-of-scope for LFG per LFP reverse-eng docs).\n- LFP lfe-core response (the \"check lfp lfe-core\" reference): Extracted the *neutral* contracts (AgentWorkState, IUlwHost, LazyCapability*, KeywordDetection, SharedSkill*, ReviewSynthesis/DecisionKernel, continuation directives, etc.) into `Lfe.AgentOs/`, `Lfe.Hooks/`, `Lfe.CodexAdapter/` (reference host only), `Lfe.CodexMcpBridge/`. LFP consumes them via ProjectReference + local NuGet; the pi-ai layer registered \"openai-codex\" + \"openai-codex-responses\" as a first-class KnownProvider. Duplication of OMO logic was deliberately avoided — LFE-core is the unification layer.\n- Current LFG \"omo-\" hierarchy (global search, excluding pure vendor/omo-standalone internals):\n  - `catalog/omo-skill-map.json`, references in doctor/mcp/tools/common.\n  - `src/runtime-ts/adapters/omo-kernel.ts` + `omo-hooks.ts` (the current OMO hook bridge).\n  - `omo-policy` string in supervision broker (spawn-adapter.ts).\n  - `omo-agent-registry-contract.json` test fixture.\n  - Many path references to `vendor/omo-standalone`.\n  - \"omo-\" prefixed tmux session helpers in dist (runtime).\n- Duplication risk identified: The current omo-kernel/hooks adapters + skill-map + policy strings are exactly the kind of \"omo- prefixed\" surface that LFE-core now unifies (post-4471). The plan's unification strategy is to make the new `lfe-core-*` adapters the primary implementation, with the existing omo-* files becoming thin legacy bridges that delegate to LFE contracts where possible, then delete the duplication over phases.\n- No other \"omo-\" product identity or installer code exists in LFG source (good — LFG correctly stayed out of OMO product surfaces).\n- Conclusion for unification: LFE-core (vendored) + LFG adaptation layer will be the single stack for the Agent OS features listed by the user. omo-standalone vendor stays only for model-core / ulw-kernel pieces not yet in LFE-core scope. This creates the post-4471 parity adaptation without a second hierarchy.\n\n(The planner performed the full PR-related file reads + LFP doc reads + \"omo-\" greps in both omo/lfp/lfg before writing the above. Implementer must repeat and append new findings to this section before Phase 0 code changes.)\n\nThe original query (\"make things like what was done on lazycodex on the omo at ~/workspace/linalab/omo check lfp lfe-core for chekcing then do implmentatuon of that on grok with this lfg then let lfg and grok be use openai-codex as a model with checking grok build manual of adding new provider\") + the line-166 feedback = **build a full OMO-like (LFE-core-powered) harnessing Agent OS that runs *inside* Grok Build via the LFG plugin**, with:\n- Main orchestrator (Sisyphus + family) defaulting to **gpt-5.5 / openai-codex** (not grok-build) while Grok remains the mandatory Oracle gate.\n- The *exact same* adaptation layer + hook + feature architecture as OMO and LFE-core: lazy LSP (on-demand MCP), shared skills, keyword triggering logic, continuation, review synthesis, ULW host contract, etc.\n- **lfe-core must be used as the implementation core**, placed as a submodule / vendored tree inside the LFG package exactly like LFP structures its `lfe-core/` (local ProjectReference equivalent + contract packages).\n\n**Why the first narrow plan was insufficient:** It only solved the prerequisite model registration. The real work is the full harnessing layer (adaptation + contracts + feature parity) so that Grok Build becomes a first-class host for the LFE-core Agent OS, just as pi-agent and LazyCodex/Codex are.\n\n**Current LFG state vs. target:**\n- Partial OMO port (vendors omo-standalone, has runtime-ts/hooks-ts/MCP-TS, codex lane in model resolution, some smoke parity).\n- Missing: lfe-core contract submodule, full adaptation layer, lazy-LSP-style on-demand backends, keyword triggering, complete shared-skill + hook parity, and production use of the gpt-5.5 orchestrator lane.\n\nLFP proved the pattern works: neutral contracts in `lfe-core/`, thin host-specific adapters in `Lfp.Integration.*`, main logic consumes the contracts. LFG will do the TS/Bun equivalent on Grok Build.\n\n---\n\n## Scope (Explicitly In / Out)\n\n**In scope (phased delivery):**\n- Phase 0: `openai-codex` as first-class model/provider (Grok manual compliant) so the orchestrator *can* use gpt-5.5.\n- Phase 1: Add `lfe-core` (neutral contracts for ULW host, hooks, Agent OS, lazy capability, review, keyword, shared skills, etc.) as a vendored submodule/tree under `plugins/lfg/` (modeled exactly on LFP's layout + reference strategy).\n- Phase 2: Grok-specific adaptation layer (new adapters/ that implement the LFE contracts) + switch main orchestrator profiles to the openai-codex/gpt-5.5 lane (Oracle still mandatory).\n- Phase 3: Feature parity (lazy LSP on-demand, shared skills, keyword triggering, full hook surface) exercised through the LFE path.\n- All docs, evidence, AGENTS contracts, smoke/self-test updated so the claim \"Grok Build now hosts a full LFE-core OMO-like Agent OS via LFG (LFP structure)\" is true and auditable.\n\n**Out of scope:**\n- Re-introducing any Python LFG code or legacy grok-harnessing.\n- Copying OMO product surfaces (codex installer, marketplace, telemetry).\n- Making LFG execute C# lfe-core at runtime (unless user later chooses sidecar); v1 uses TS projections of the contracts (LFG stays pure Bun/TS).\n- Replacing Grok's native TUI — LFG extends it.\n\n---\n\n## Recommended Approach (LFP Structure on Grok)\n\n**Target architecture (LFE-core as unification layer — remove OMO duplication per @plan.md:14 feedback):**\nLFE-core (the neutral contracts extracted post-OMO PR #4471, as proven in LFP) becomes the single source of truth for the Agent OS (ULW host, hooks, lazy capability, keyword detection, shared skills, review synthesis, continuation, AgentWorkState, etc.).\n\nLFG's job is the thin Grok-specific adaptation layer on top (exactly as LFP's `Lfp.Integration.*` sits on `lfe-core/`).\n\n**Explicit goal:** Audit and *remove* \"omo-\" prefixed duplication in the code hierarchy wherever LFE-core now provides the unifying contract. The omo-standalone vendor remains only for the parts not yet unified under LFE-core (or as a temporary bridge during migration). No new parallel \"omo-\" stacks are created.\n\n**Target layout (LFP pattern + unification):**\n```\nLFG (Grok plugin)\n├── vendor/lfe-core/                 # THE unifying core (contracts from LFP lfe-core, post PR 4471 learnings)\n├── vendor-links/lfe-core/           # re-exports\n├── src/runtime-ts/\n│   ├── adapters/\n│   │   └── lfe-core-*.ts            # The *only* new adaptation layer (Grok-specific impl of LFE contracts)\n│   │   └── omo-*.ts (legacy bridge only; slated for removal as LFE unification progresses)\n│   └── services/ (model-resolution defaults orchestrator to openai-codex/gpt-5.5 via LFE path)\n└── docs/LFE-CORE-UNIFICATION.md (the migration/unification plan + duplication audit results)\n```\n\n**Phasing (unification first, remove duplication, land value safely):**\n1. **Phase 0 (small, unblocks everything):** openai-codex provider/model (Grok manual compliant). This is the prerequisite so the main orchestrator *can* use gpt-5.5.\n2. **Phase 1 (Unification foundation):** Add lfe-core as the submodule/unifying core (per LFP layout). Full duplication audit of all \"omo-\" code in LFG + vendor. Document in LFE-CORE-UNIFICATION.md exactly which OMO surfaces will be removed in favor of LFE-core contracts. Update ADAPTER-BOUND-DISPOSITION.md.\n3. **Phase 2:** Thin Grok adaptation layer on *LFE-core* (not a second parallel stack) + switch orchestrator profiles to openai-codex/gpt-5.5. Existing pure-TS paths become the \"legacy bridge\" and are marked for later removal.\n4. **Phase 3:** Implement the LFE-core-powered features (lazy LSP, shared skills, keyword triggering, full hooks, continuation) as the *only* path. Begin removing duplicated \"omo-\" logic.\n5. **Phase 4:** Full evidence on Grok Build TUI + release gates. The hierarchy is now LFE-core (core) → LFG adaptation (Grok host) with no OMO duplication for the unified surfaces.\n\n**Reuse (heavy):**\n- Existing vendor/ + vendor-links/ pattern for omo-standalone (copy for lfe-core).\n- Current runtime-ts, hooks-ts, MCP-TS, spawn-adapter, model-resolution, openai-compatible-dispatch.\n- Grok custom model mechanism (exactly as documented in `~/.grok/docs/user-guide/11-custom-models.md`).\n- LFP docs as spec (lfe-core-integration.md, lazycodex-reverse-engineering.md, architecture.md) — read-only.\n\n**Verification:** Every phase ends with full `bun test` matrix + `self-test.ts` + at least one manual Grok TUI trace exercising the new surface. Grok Oracle gate never bypassed.\n\n---\n\n## Critical Files (High-Level)\n\n**Edit (existing LFG):**\n- `plugins/lfg/src/runtime-ts/services/model-resolution.ts` + tests (Phase 0 + 2)\n- `plugins/lfg/src/runtime-ts/commands/provider.*` + common.ts\n- `plugins/lfg/config/grok-config-template.toml` (add [model.openai-codex] with responses)\n- `plugins/lfg/src/runtime-ts/services/openai-compatible-dispatch.ts` + spawn-adapter.ts\n- `plugins/lfg/docs/ARCHITECTURE.md`, `SMOKE.md`, `skills/provider/SKILL.md`\n- `plugins/lfg/bin/self-test.ts`, smoke-ts checks, fixtures\n- All 5 `AGENTS.md` (structural updates only)\n- `plugins/lfg/src/runtime-ts/adapters/*` (add lfe-core siblings)\n\n**New:**\n- `plugins/lfg/vendor/lfe-core/` (contracts)\n- `plugins/lfg/vendor-links/lfe-core/`\n- `plugins/lfg/src/runtime-ts/adapters/lfe-core-*.ts` (several)\n- `plugins/lfg/docs/LFE-CORE-INTEGRATION.md`\n- `plugins/lfg/docs/lazycodex-lfe-parity-audit.md` (the \"check lfp lfe-core\" evidence)\n\n**Read-only during work (and the implementer MUST re-audit these before every phase):**\n- OMO PR #4471 in full (all files, tests, docs, omo-codex package) — this is the primary source of \"what was done on lazycodex\".\n- Everything in `~/workspace/linalab/lfp/lfe-core/` and its docs (lazycodex-reverse-engineering.md, lfe-core-integration.md, lfe-core-agent-os-consumption.md, the actual Lfe.CodexAdapter / Lfe.CodexMcpBridge / Lfe.AgentOs sources) — the \"check lfp lfe-core for checking\" reference.\n- `~/.grok/docs/user-guide/11-custom-models.md` (exact TOML + api_backend rules for openai-codex).\n- Current LFG vendor/omo-standalone + runtime-ts + hooks-ts (to find the exact insertion points for the LFE adaptation layer).\n\nThe plan author re-audited the above (via list_dir, multiple greps, and targeted reads of the key PR-related paths and LFP docs) before finalizing this version. The implementer must repeat the audit.\n\n---\n\n## Verification (Phased, Strict)\n\n**Phase 0 gate (openai-codex model):** `lfg provider add --kind openai-codex`, spawn with that provider, Grok config.toml section per manual, all tests + self-test green, docs updated.\n\n**Phase 1 gate:** lfe-core tree present and importable; no regression in OMO vendor; integration doc written.\n\n**Phase 2+ gate:** End-to-end manual run on Grok TUI where Sisyphus/Hephaestus etc. resolve to gpt-5.5 via LFE-core adaptation layer, lazy-LSP / keyword / shared-skill features fire, evidence recorded, Oracle still required. Full test matrix + self-test pass. All AGENTS.md complete.\n\n**Global:** Only files under `plugins/lfg/` change. No Python, no weakening of existing contracts or Oracle gate.\n\n","total_lines":363,"mtime_at_read":{"Known":1780078891387},"content_hash":14295602953664408115}}},"_meta":{"totalTokens":88408,"eventId":"019e7466-d18d-71b0-9b1f-6599dc5defbc-45352","agentTimestampMs":1780087108357,"promptId":"task-completed-call-94d148c9-4a45-4451-9d26-fc8b3bb949cb-20","streamStartMs":1780087106722,"turnStartMs":1780087097614,"updateType":"ToolCallUpdate","updateParams":{"toolCallId":"call-29eb3720-555f-44d3-8b22-2a482d8b3d62-50","status":"Completed"}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e7466-d18d-71b0-9b1f-6599dc5defbc/summary.json:6:  "session_summary": "Implement LazyCodex Checks on Grok LFG Using OpenAI-Codex Provider",
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e7466-d18d-71b0-9b1f-6599dc5defbc/summary.json:23:  "generated_title": "Implement LazyCodex Checks on Grok LFG Using OpenAI-Codex Provider",
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e7466-d18d-71b0-9b1f-6599dc5defbc/terminal/call-fb0f4374-264f-4e0f-8da9-9a99f7cc61d0-51.log:15:- Scope limit: only model/provider surface, or also a minimal "Codex host session adapter" in LFG hooks for when Grok is invoked from a LazyCodex/Codex environment? (Rec: model surface only for this ticket; second slice if user confirms.)
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86b4-2349-7e21-92db-e66412d91bc1/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86b4-2349-7e21-92db-e66412d91bc1/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["cua-driver","0-1-0-ff47fdd7:ulw-loop","0-1-0-ff47fdd7:review-work","0-1-0-ff47fdd7:frontend-ui-ux","0-1-0-ff47fdd7:start-work","0-1-0-ff47fdd7:rules","help","review","best-of-n","0-1-0-ff47fdd7:ulw-plan","github-yeet","k-skill-setup","pptx","check-work","implement","0-1-0-ff47fdd7:refactor","0-1-0-ff47fdd7:init-deep","docx","korean-law-search","xlsx","execute-plan","pr-babysit","0-1-0-ff47fdd7:remove-ai-slops","create-skill","0-1-0-ff47fdd7:comment-checker","0-1-0-ff47fdd7:lsp","mfds-food-safety","lina-decision-protocol","0-1-0-ff47fdd7:programming","design","0-1-0-ff47fdd7:debugging"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86b4-2349-7e21-92db-e66412d91bc1/updates.jsonl:1:{"timestamp":1780376413,"method":"session/update","params":{"sessionId":"019e86b4-2349-7e21-92db-e66412d91bc1","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4195,"eventId":"019e86b4-2349-7e21-92db-e66412d91bc1-0","agentTimestampMs":1780376413541,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86b4-2349-7e21-92db-e66412d91bc1/updates.jsonl:3:{"timestamp":1780376414,"method":"session/update","params":{"sessionId":"019e86b4-2349-7e21-92db-e66412d91bc1","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e86b4-2349-7e21-92db-e66412d91bc1-1","agentTimestampMs":1780376413542,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bff-7341-949c-b9678d4a7b2e/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bff-7341-949c-b9678d4a7b2e/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["0-1-0-ff47fdd7:start-work","0-1-0-ff47fdd7:ulw-loop","execute-plan","help","implement","0-1-0-ff47fdd7:ulw-plan","xlsx","check-work","0-1-0-ff47fdd7:init-deep","github-yeet","0-1-0-ff47fdd7:frontend-ui-ux","0-1-0-ff47fdd7:lsp","docx","mfds-food-safety","korean-law-search","0-1-0-ff47fdd7:programming","review","lina-decision-protocol","0-1-0-ff47fdd7:rules","0-1-0-ff47fdd7:review-work","0-1-0-ff47fdd7:refactor","0-1-0-ff47fdd7:comment-checker","pr-babysit","pptx","0-1-0-ff47fdd7:debugging","create-skill","k-skill-setup","0-1-0-ff47fdd7:remove-ai-slops","best-of-n","cua-driver","design"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bff-7341-949c-b9678d4a7b2e/updates.jsonl:1:{"timestamp":1780386786,"method":"session/update","params":{"sessionId":"019e8752-6bff-7341-949c-b9678d4a7b2e","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4256,"eventId":"019e8752-6bff-7341-949c-b9678d4a7b2e-0","agentTimestampMs":1780386786995,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bff-7341-949c-b9678d4a7b2e/updates.jsonl:3:{"timestamp":1780386787,"method":"session/update","params":{"sessionId":"019e8752-6bff-7341-949c-b9678d4a7b2e","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e8752-6bff-7341-949c-b9678d4a7b2e-1","agentTimestampMs":1780386786996,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bff-7341-949c-b9678d4a7b2e/updates.jsonl:9:{"timestamp":1780386791,"method":"session/update","params":{"sessionId":"019e8752-6bff-7341-949c-b9678d4a7b2e","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video"]}},"_meta":{"totalTokens":16694,"eventId":"019e8752-6bff-7341-949c-b9678d4a7b2e-14","agentTimestampMs":1780386791595,"promptId":"336f0981-33e7-4ef0-b751-f7f1abeb16a9","streamStartMs":1780386789239,"turnStartMs":1780386787971,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n- cancel: Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md\n- agent-browser: Automates browser interactions for web testing, form filling, screenshots, and data extraction\n  Use when: Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md\n- pre-publish-review: Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-r…\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md\n- remove-deadcode: Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md\n- github-triage: Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md\n- ai-slop-remover: Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md\n- plan: Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md\n- work-with-pr: Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge\n  Use when: Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md\n- team-mode: Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md\n- doctor: Diagnose and fix LFG plugin state, schema, and provider configuration issues.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md\n- get-unpublished-changes: Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md\n- setup: Install/sync the LFG Grok plugin and configure provider/model metadata.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md\n- provider: List, show, or add provider/model metadata without storing secret values.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md\n- ralph: Durable self-referential development loop that continues until the objective is met.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md\n- omomomo: Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md\n- playwright: MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md\n- publish: Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md\n- git-master: MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md\n- team: Create and manage durable OMO-style agent teams with tmux-backed local observability.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md\n- lfg: Use this skill when validating the `islee23520/lfg` plugin package.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md\n- start-work: Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md\n- hyperplan: Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization\n  Use when: Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md\n- dev-browser: Browser automation with persistent page state\n  Use when: Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md\n- worker: Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md\n- ulw: Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/chat_history.jsonl:15:{"type":"tool_result","tool_call_id":"call-0d26feb2-663c-468a-a265-d13c3e3ed978-3","content":"1→# Grok Build Custom Models Configuration Template\n2→# Copy this file to config.toml and replace {{PLACEHOLDER}} values with your actual configuration\n3→\n4→[model.litellm]\n5→# Base URL for the litellm-compatible API endpoint\n6→# Example: \"https://api.litellm.ai/v1\"\n7→base_url = \"{{PLACEHOLDER}}\"\n8→\n9→# Environment variable name containing the API key\n10→# The runtime will read the actual key from this environment variable\n11→# Example: \"LITELLM_API_KEY\"\n12→env_key = \"{{PLACEHOLDER}}\"\n13→\n14→# Model identifier to use for this provider\n15→# Example: \"grok-1\" or \"claude-3-opus\"\n16→model = \"{{PLACEHOLDER}}\"\n17→\n18→[model.openai-compatible]\n19→# Base URL for an OpenAI-compatible API endpoint\n20→# Example: \"https://api.openai.com/v1\"\n21→base_url = \"{{PLACEHOLDER}}\"\n22→\n23→# Environment variable name containing the API key\n24→# Example: \"OPENAI_API_KEY\"\n25→env_key = \"{{PLACEHOLDER}}\"\n26→\n27→# Model identifier to use for this provider\n28→# Example: \"gpt-4-turbo\" or \"gpt-4\"\n29→model = \"{{PLACEHOLDER}}\"\n30→\n31→# openai-codex (gpt-5.5 / codex deep-specialist lane for main orchestrator)\n32→# Registered per Grok Build manual (11-custom-models.md) + LFP pi-ai \"openai-codex-responses\" pattern.\n33→# Use api_backend = \"responses\" for the Responses API shape (matches LFE-core / post-4471 codex expectations).\n34→# This is the primary execution provider for Sisyphus-family agents in the LazyCodex-compatible MVP.\n35→[model.openai-codex]\n36→model = \"openai-codex\"\n37→# base_url points at your codex-compatible or OpenAI-compatible endpoint that serves *-codex / gpt-5.x-codex models\n38→base_url = \"{{PLACEHOLDER}}\"\n39→env_key = \"{{PLACEHOLDER}}\"\n40→api_backend = \"responses\"\n41→auth_scheme = \"bearer\"\n42→# Optional: context_window, temperature, etc. per Grok manual\n43→\n44→# Codex OAuth option:\n45→# 1. Verify readiness first:\n46→#      lfg --json auth codex-oauth status\n47→# 2. If your Grok host should obtain auth through the Codex app login, configure the\n48→#    external provider explicitly. This is global Grok auth, not a per-model field.\n49→#\n50→# [auth]\n51→# auth_provider_command = \"bun /absolute/path/to/plugins/lfg/bin/codex-oauth-provider.ts --json-token\"\n52→# auth_provider_label = \"Codex OAuth\"\n53→\n54→[models]\n55→# Default model identifier to use when no model is explicitly specified\n56→# Should match one of the model sections above (e.g., \"grok\", \"litellm\", \"openai-compatible\")\n57→default = \"grok\"\n58→"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/chat_history.jsonl:25:{"type":"tool_result","tool_call_id":"call-3e0ddbc2-725c-4e64-9e48-366536d984c7-9","content":"<workspace_result workspace_path=\"/Users/ilseoblee/workspace/linalab/lfg\">\nFound at least 27 matching lines\n/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/lazycodex-grok-inspect-after-enable.json\n1414:        \"path\": \"/Users/ilseoblee/.grok/config.toml\"\n1450:    \"userPath\": \"/Users/ilseoblee/.grok/config.toml\",\n\n/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/lazycodex-grok-inspect.json\n1444:        \"path\": \"/Users/ilseoblee/.grok/config.toml\"\n1450:    \"userPath\": \"/Users/ilseoblee/.grok/config.toml\",\n\n/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/task-8-review-diff.txt\n24: - `openai-codex` is now a first-class approved provider + model (with Grok manual-compliant `[model.openai-codex]` template section using `api_backend = \"responses\"`).\n40:@@ -54,7 +54,7 @@ Grok-specific pieces (TUI integration, spawn envelope, Oracle gate, config.toml\n443:   // [model.openai-codex] section from the config template (responses backend per Grok manual + LFP parity).\n\n/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/task-8-final.txt\n286:✓ grok-config-template.toml > should contain required fields in [model.litellm]\n288:✓ grok-config-template.toml > should contain required fields in [model.openai-co\n\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js\n9351:  const configPath = join6(resolveLfgEnv().root, \"config\", \"config.toml\");\n10426:  const configPath = join17(home, \".grok\", \"config.toml\");\n10500:      \"You can further customize ~/.grok/agents/lina.md or add [subagents.roles.lina] in config.toml\"\n10506:  const configPath = join18(homeDir(input.processEnv), \".grok\", \"config.toml\");\n10521:  const configPath = join18(home, \".grok\", \"config.toml\");\n10746:  { workflow: \"grok_manual_gates\", status: \"manual-gated\", surface: \"host-native Grok child spawning and live external model execution\", evidence: \"~/.grok/docs/user-guide/09-plugins.md; ~/.grok/docs/user-guide/11-custom-models.md; docs/SMOKE.md manual gate notes\" }\n10870:    \"optional: add [model.openai-codex] from plugins/lfg/config/grok-config-template.toml when the host should route Codex-class external models\",\n10882:      globalConfigCaveat: \"LFG reports Grok setup steps only; it does not write ~/.grok/config.toml or run grok plugin install.\",\n10890:        optionalModel: `[model.openai-codex]\n\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js.map\n159:    \"import { existsSync, readFileSync } from \\\"node:fs\\\"\\nimport { join } from \\\"node:path\\\"\\nimport { resolveModel as resolveCoreModel, resolveModelWithFallback as resolveCoreModelWithFallback } from \\\"../omo-compat/model-core\\\"\\nimport { resolveLfgEnv, type LfgEnv } from \\\"../foundation/env\\\"\\nimport { loadOmoAgentRegistry, type OmoAgent } from \\\"./agent-registry\\\"\\nimport { isLfeCoreBoulderActive } from \\\"../adapters/lfe-core-hooks\\\"\\nimport { getLfeCoreModelProfileSuggestion } from \\\"../adapters/lfe-core-ulw\\\"\\nimport { OPENAI_CODEX_DEFAULT_MODEL } from \\\"./openai-codex\\\"\\nimport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n} from \\\"./provider-catalog\\\"\\n\\nexport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n}\\nexport type ReasoningLe [... truncated (1035 chars total)]\n179:    \"import { join, resolve } from \\\"node:path\\\"\\nimport { bootstrapState, ensureStateSchema, stateSchemaPath } from \\\"../foundation/state-schema\\\"\\nimport { commandEnv, directoryExists, executablePath, pathExists, readCatalogSkillCount, readJsonObject, type CommandContext, type JsonObject } from \\\"./common\\\"\\nimport { readCodexOAuthStatus } from \\\"../services/codex-oauth\\\"\\nimport { getLazyCodexCapabilities } from \\\"../services/lazy-capabilities\\\"\\n\\nexport async function doctor(context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const checks: JsonObject[] = []\\n  const add = (name: string, ok: boolean, evidence: string, required = true): void => { checks.push({ name, ok, required, evidence }) }\\n  const manifest = join(env.root, \\\".grok-plugin\\\", \\\"plugin.json\\\")\\n  add(\\\"grok_manifest\\\", await pathExists(manifest), manifest)\\n  const mcpConfig = join(env.root, \\\".mcp.json\\\")\\n  add(\\\"mcp_config\\\", await pathExis [... truncated (1035 chars total)]\n181:    \"import { access, copyFile, mkdir, readFile, readdir, writeFile } from \\\"node:fs/promises\\\"\\nimport { dirname, join } from \\\"node:path\\\"\\nimport type { LfgEnv } from \\\"../foundation/env\\\"\\n\\nexport type NativeAgentEnv = NodeJS.ProcessEnv\\n\\nexport type NativeAgentResult = Record<string, unknown>\\n\\nexport async function registerNativeAgent(input: { readonly agentId: string; readonly force: boolean; readonly env: LfgEnv; readonly processEnv: NativeAgentEnv }): Promise<NativeAgentResult> {\\n  const candidates = sourceCandidates(input.agentId, input.env, input.processEnv)\\n  const sourcePath = await firstExistingPath(candidates)\\n  if (!sourcePath) {\\n    return {\\n      ok: false,\\n      error: `Could not find source definition for agent \\\"${input.agentId}\\\"`,\\n      tried: candidates,\\n      hint: \\\"Run from within the lfg plugin or set GROK_PLUGIN_ROOT\\\",\\n    }\\n  }\\n\\n  const destDir = join(homeDir(input.processEnv), \\\".grok\\\", \\\"agents\\\")\\n  const destPath = join(destDir, `$ [... truncated (1035 chars total)]\n184:    \"export type LazycodexWorkflowStatus = \\\"implemented\\\" | \\\"documented-equivalent\\\" | \\\"manual-gated\\\" | \\\"out-of-scope\\\"\\n\\nexport type LazycodexWorkflowRow = {\\n  readonly workflow: string\\n  readonly status: LazycodexWorkflowStatus\\n  readonly surface: string\\n  readonly evidence: string\\n}\\n\\nexport type LazycodexStatusResult = {\\n  readonly ok: true\\n  readonly status: \\\"ok\\\"\\n  readonly command: \\\"lazycodex status\\\"\\n  readonly source: string\\n  readonly rows: LazycodexWorkflowRow[]\\n  readonly summary: {\\n    readonly total: number\\n    readonly implemented: number\\n    readonly documentedEquivalent: number\\n    readonly manualGated: number\\n    readonly outOfScope: number\\n  }\\n}\\n\\nconst LAZYCODEx_ROWS = [\\n  { workflow: \\\"install_bootstrap\\\", status: \\\"documented-equivalent\\\", surface: \\\"lfg setup check/show/install-plan plus Grok plugin manifest\\\", evidence: \\\"plugins/lfg/bin/lfg.ts setup; plugins/lfg/.claude-plugin/plugin.json; docs/SMOKE.md\\\" },\\n  { workflow: \\\"pro [... truncated (1035 chars total)]\n187:    \"import { join } from \\\"node:path\\\"\\nimport { bootstrapState } from \\\"../foundation/state-schema\\\"\\nimport { asRecord, commandEnv, copyTree, detectRepo, pathExists, readJson, readProviderState, setupPath, utcNow, writeJson, type CommandContext, type JsonObject } from \\\"./common\\\"\\n\\nexport type SetupOptions = { pluginDir?: string; dryRun?: boolean }\\nexport type SetupInstallPlanOptions = { marketplace?: string }\\n\\nexport async function setup(options: SetupOptions = {}, context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const dest = options.pluginDir ?? process.env.LFG_PLUGIN_DEST ?? join(process.env.HOME ?? \\\"\\\", \\\".grok\\\", \\\"plugins\\\", \\\"lfg\\\")\\n  const dryRun = Boolean(options.dryRun)\\n  const providerState = await readProviderState(env)\\n  if (!dryRun) await copyTree(env.root, dest)\\n  const record: JsonObject = { ok: true, dryRun, installed: !dryRun, plugin: { source: env.root, dest, exists: await pathExi [... truncated (1035 chars total)]\n\n/Users/ilseoblee/workspace/linalab/lfg/./plans/lazycodex-grokbuild-native-openai-sdk.md\n22:- Default applied: \"native Grok Build support\" means Grok plugin/skill/agent/hook/MCP/custom-model surfaces are discoverable and test-backed; real host child spawning remains manual-gated unless a Grok transcript proves it.\n219:  - `/Users/ilseoblee/.grok/docs/user-guide/11-custom-models.md:56` - Grok API backend options.\n220:  - `/Users/ilseoblee/.grok/docs/user-guide/11-custom-models.md:68` - Grok auth scheme options.\n286:  **What to do**: Update setup/status docs and deterministic setup output so `lfg setup check` or `lfg setup install-plan` reports the exact Grok steps for LazyCodex-compatible plugin use: plugin install/trust, skill discovery, MCP discovery, optional `[auth] auth_provider_command`, optional `[model.openai-codex]`, and manual Grok gates. If `setup` command output is already the intended place, extend it; otherwise add a `lazycodex` section to `doctor`.\n</workspace_result>"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/chat_history.jsonl:27:{"type":"tool_result","tool_call_id":"call-3e0ddbc2-725c-4e64-9e48-366536d984c7-11","content":"<workspace_result workspace_path=\"/Users/ilseoblee/workspace/linalab/lfg\">\nFound 21 matching lines\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js\n7334:  constructor({ baseURL = readEnv(\"OPENAI_BASE_URL\"), apiKey = readEnv(\"OPENAI_API_KEY\") ?? null, adminAPIKey = readEnv(\"OPENAI_ADMIN_KEY\") ?? null, organization = readEnv(\"OPENAI_ORG_ID\") ?? null, project = readEnv(\"OPENAI_PROJECT_ID\") ?? null, webhookSecret = readEnv(\"OPENAI_WEBHOOK_SECRET\") ?? null, workloadIdentity, ...opts } = {}) {\n7955:    constructor({ baseURL = readEnv(\"OPENAI_BASE_URL\"), apiKey = readEnv(\"AZURE_OPENAI_API_KEY\"), apiVersion = readEnv(\"OPENAI_API_VERSION\"), endpoint, deployment, azureADTokenProvider, dangerouslyAllowBrowser, ...opts } = {}) {\n8068:  const baseUrl = options.baseUrl ?? OFFICIAL_OPENAI_BASE_URL;\n8073:  if (baseUrl !== OFFICIAL_OPENAI_BASE_URL) {\n8133:var CODEX_OAUTH_PROVIDER = \"openai-codex\", OFFICIAL_OPENAI_BASE_URL = \"https://api.openai.com/v1\";\n\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js.map\n147:    \"// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.\\nvar _OpenAI_instances, _a, _OpenAI_encoder, _OpenAI_baseURLOverridden;\\nimport { __classPrivateFieldGet, __classPrivateFieldSet } from \\\"./internal/tslib.mjs\\\";\\nimport { uuid4 } from \\\"./internal/utils/uuid.mjs\\\";\\nimport { validatePositiveInteger, isAbsoluteURL, safeJSON } from \\\"./internal/utils/values.mjs\\\";\\nimport { sleep } from \\\"./internal/utils/sleep.mjs\\\";\\nimport { castToError, isAbortError } from \\\"./internal/errors.mjs\\\";\\nimport { getPlatformHeaders } from \\\"./internal/detect-platform.mjs\\\";\\nimport * as Shims from \\\"./internal/shims.mjs\\\";\\nimport * as Opts from \\\"./internal/request-options.mjs\\\";\\nimport { stringifyQuery } from \\\"./internal/utils/query.mjs\\\";\\nimport { VERSION } from \\\"./version.mjs\\\";\\nimport * as Errors from \\\"./core/error.mjs\\\";\\nimport * as Pagination from \\\"./core/pagination.mjs\\\";\\nimport { WorkloadIdentityAuth } from \\\"./auth/workload-identity-auth.mj [... truncated (1035 chars total)]\n148:    \"import { buildHeaders } from \\\"./internal/headers.mjs\\\";\\nimport * as Errors from \\\"./error.mjs\\\";\\nimport { isObj, readEnv } from \\\"./internal/utils.mjs\\\";\\nimport { OpenAI } from \\\"./client.mjs\\\";\\n/** API Client for interfacing with the Azure OpenAI API. */\\nexport class AzureOpenAI extends OpenAI {\\n    /**\\n     * API Client for interfacing with the Azure OpenAI API.\\n     *\\n     * @param {string | undefined} [opts.apiVersion=process.env['OPENAI_API_VERSION'] ?? undefined]\\n     * @param {string | undefined} [opts.endpoint=process.env['AZURE_OPENAI_ENDPOINT'] ?? undefined] - Your Azure endpoint, including the resource, e.g. `https://example-resource.azure.openai.com/`\\n     * @param {string | undefined} [opts.apiKey=process.env['AZURE_OPENAI_API_KEY'] ?? undefined]\\n     * @param {string | undefined} opts.deployment - A model deployment, if given, sets the base client URL to include `/deployments/{deployment}`.\\n     * @param {string | null | undefined} [opts.organizatio [... truncated (1035 chars total)]\n150:    \"import { buildCodexOAuthProviderPayload, readCodexOAuthStatus, type CodexOAuthStatus } from \\\"./codex-oauth\\\"\\n\\nexport type OpenAISdkProbeMode = \\\"dry-run\\\" | \\\"live\\\"\\nexport type OpenAISdkProbeFetch = (input: Parameters<typeof globalThis.fetch>[0], init?: Parameters<typeof globalThis.fetch>[1]) => Promise<Response>\\n\\nexport type OpenAISdkProbeOptions = {\\n  readonly authJsonPath?: string\\n  readonly baseUrl?: string\\n  readonly env?: Readonly<Record<string, string | undefined>>\\n  readonly fetch?: OpenAISdkProbeFetch\\n  readonly input?: string\\n  readonly mode?: OpenAISdkProbeMode\\n  readonly model: string\\n  readonly provider?: string\\n  readonly timeoutMs?: number\\n}\\n\\nexport type OpenAISdkProbeResult = {\\n  readonly ok: boolean\\n  readonly status: \\\"dry-run\\\" | \\\"ok\\\" | \\\"blocked\\\" | \\\"unavailable\\\" | \\\"failed\\\"\\n  readonly provider: string\\n  readonly model: string\\n  readonly baseUrl: string\\n  readonly mode: OpenAISdkProbeMode\\n  readonly networkCalled: boolean\\n  r [... truncated (1035 chars total)]\n179:    \"import { join, resolve } from \\\"node:path\\\"\\nimport { bootstrapState, ensureStateSchema, stateSchemaPath } from \\\"../foundation/state-schema\\\"\\nimport { commandEnv, directoryExists, executablePath, pathExists, readCatalogSkillCount, readJsonObject, type CommandContext, type JsonObject } from \\\"./common\\\"\\nimport { readCodexOAuthStatus } from \\\"../services/codex-oauth\\\"\\nimport { getLazyCodexCapabilities } from \\\"../services/lazy-capabilities\\\"\\n\\nexport async function doctor(context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const checks: JsonObject[] = []\\n  const add = (name: string, ok: boolean, evidence: string, required = true): void => { checks.push({ name, ok, required, evidence }) }\\n  const manifest = join(env.root, \\\".grok-plugin\\\", \\\"plugin.json\\\")\\n  add(\\\"grok_manifest\\\", await pathExists(manifest), manifest)\\n  const mcpConfig = join(env.root, \\\".mcp.json\\\")\\n  add(\\\"mcp_config\\\", await pathExis [... truncated (1035 chars total)]\n187:    \"import { join } from \\\"node:path\\\"\\nimport { bootstrapState } from \\\"../foundation/state-schema\\\"\\nimport { asRecord, commandEnv, copyTree, detectRepo, pathExists, readJson, readProviderState, setupPath, utcNow, writeJson, type CommandContext, type JsonObject } from \\\"./common\\\"\\n\\nexport type SetupOptions = { pluginDir?: string; dryRun?: boolean }\\nexport type SetupInstallPlanOptions = { marketplace?: string }\\n\\nexport async function setup(options: SetupOptions = {}, context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const dest = options.pluginDir ?? process.env.LFG_PLUGIN_DEST ?? join(process.env.HOME ?? \\\"\\\", \\\".grok\\\", \\\"plugins\\\", \\\"lfg\\\")\\n  const dryRun = Boolean(options.dryRun)\\n  const providerState = await readProviderState(env)\\n  if (!dryRun) await copyTree(env.root, dest)\\n  const record: JsonObject = { ok: true, dryRun, installed: !dryRun, plugin: { source: env.root, dest, exists: await pathExi [... truncated (1035 chars total)]\n\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/src/runtime-ts/services/openai-sdk-probe.ts\n47:const OFFICIAL_OPENAI_BASE_URL = \"https://api.openai.com/v1\"\n51:  const baseUrl = options.baseUrl ?? OFFICIAL_OPENAI_BASE_URL\n56:  if (baseUrl !== OFFICIAL_OPENAI_BASE_URL) {\n\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/README.md\n61:Grok's `auth_provider_command` is a global `[auth]` setting, not a per-model field. Configure it explicitly in `~/.grok/config.toml` or `GROK_AUTH_PROVIDER_COMMAND`; LFG does not mutate that file automatically. `--json-token` emits only the access token JSON payload and never forwards the cached refresh token.\n\n/Users/ilseoblee/workspace/linalab/lfg/./README.md\n138:The Grok manual exposes `auth_provider_command` as a global `[auth]` setting or `GROK_AUTH_PROVIDER_COMMAND` environment variable. LFG does not mutate `~/.grok/config.toml`; if you wire Codex OAuth into Grok, configure it explicitly and verify it in the host. `--json-token` emits only `{\"access_token\":\"...\"}` on stdout so the host never receives the refresh token from the Codex app cache:\n\n/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/task-7-docs-qa.txt\n55:README.md:138:The Grok manual exposes `auth_provider_command` as a global `[auth]` setting or `GROK_AUTH_PROVIDER_COMMAND` environment variable. LFG does not mutate `~/.grok/config.toml`; if you wire Codex OAuth into Grok, configure it explicitly and verify it in the host:\n82:plugins/lfg/README.md:61:Grok's `auth_provider_command` is a global `[auth]` setting, not a per-model field. Configure it explicitly in `~/.grok/config.toml` or `GROK_AUTH_PROVIDER_COMMAND`; LFG does not mutate that file automatically.\n\n/Users/ilseoblee/workspace/linalab/lfg/./plans/lazycodex-grokbuild-codex-oauth.md\n223:  **What to do**: Update `plugins/lfg/skills/lfg`, `provider`, `lsp`, `ast-grep`, root README, plugin README, architecture docs, smoke docs, and config template. Document the combined story: LazyCodex-style Grok Build plugin, `lina` public surface, lazy MCP tools, and Codex OAuth for `openai-codex`. Include global Grok auth caveat for `auth_provider_command`.\n\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/skills/provider/SKILL.md\n29:- **Grok auth caveat**: Grok's `auth_provider_command` is global `[auth]` configuration or `GROK_AUTH_PROVIDER_COMMAND`, not provider metadata stored by `/provider add`. LFG does not edit `~/.grok/config.toml` automatically.\n\n/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/config/grok-config-template.toml\n37:# base_url points at your codex-compatible or OpenAI-compatible endpoint that serves *-codex / gpt-5.x-codex models\n\n/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-codex-oauth-1-baseline.txt\n1174:# base_url points at your codex-compatible or OpenAI-compatible endpoint that se\n</workspace_result>"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/chat_history.jsonl:34:{"type":"tool_result","tool_call_id":"call-37e12d3b-8559-4d12-b310-2a5c042f68d6-14","content":"40→  add(\"default_launcher\", true, `lfg (effective=${env.launcher})`, false)\n41→  const schema = await ensureStateSchema(env)\n42→  add(\"state_schema\", schema.version === 2 && await pathExists(stateSchemaPath(env)), `${stateSchemaPath(env)} version=${schema.version}`, true)\n43→  const tsSource = join(env.root, \"src\", \"runtime-ts\", \"index.ts\")\n44→  add(\"typescript_runtime\", await pathExists(tsSource), `${tsSource} test=bun test plugins/lfg/src/runtime-ts/index.test.ts`, true)\n45→  const providers = await teamProviderMatrix()\n46→  const available = providers.filter((provider) => provider.available).map((provider) => provider.provider)\n47→  add(\"team_provider_commands\", true, `available=${available.join(\",\")} providers=${providers.map((provider) => provider.provider).join(\",\")}`, false)\n48→  const bridge = await hookBridgeStatus(env.root)\n49→  add(\"global_hook_bridge\", Boolean(bridge.ok), `installed=${String(bridge.installed)} valid=${String(bridge.valid)} config=${String(bridge.config)}`, false)\n50→  const codexOAuth = await readCodexOAuthStatus({ env: context.processEnv ?? process.env })\n51→  add(\"codex_oauth\", codexOAuth.ok, `status=${codexOAuth.status} auth=${codexOAuth.authPath} has_refresh=${String(codexOAuth.hasRefreshToken)}`, false)\n52→\n53→  // Lina surface agent as default check (Lina mode override)\n54→  const home = process.env.HOME || process.env.USERPROFILE || \"\"\n55→  const configPath = join(home, \".grok\", \"config.toml\")\n56→  const agentsDir = join(home, \".grok\", \"agents\")\n57→  const hasLinaDef = await pathExists(join(agentsDir, \"lina.md\"))\n58→  let linaIsDefault = false\n59→  try {\n60→    const fs = await import(\"node:fs/promises\")\n61→    const cfg = await fs.readFile(configPath, \"utf8\")\n62→    linaIsDefault = /default\\s*=\\s*\"lina\"/.test(cfg)\n63→  } catch {}\n64→  const linaModeOk = hasLinaDef && linaIsDefault\n65→  add(\"lina_surface_default\", linaModeOk, `lina.md=${hasLinaDef} config_default=lina=${linaIsDefault} (run 'lfg agents register-native lina && lfg agents set-default lina' to activate full Lina mode)`, false)\n66→\n67→  const failedRequired = checks.filter((check) => check.required === true && check.ok !== true)\n68→  const warnings = checks.filter((check) => check.required !== true && check.ok !== true)\n69→  return { ok: failedRequired.length === 0, status: failedRequired.length === 0 ? \"pass\" : \"fail\", pluginRoot: env.root, pluginData: env.data, checks, capabilities: getLazyCodexCapabilities(), failedRequired, warnings }\n70→}\n71→\n72→export async function doctorStateSchemaCheck(context: CommandContext = {}): Promise<JsonObject> {\n73→  const env = commandEnv(context)\n74→  const schema = await bootstrapState(env)\n75→  return { ok: true, status: \"pass\", operation: \"doctor_state_schema_check\", schema, stateRoots: { state: env.stateDir, runs: env.runsDir, plans: env.plansDir, boulder: join(env.data, \"boulder\"), notepads: join(env.data, \"notepads\"), mailbox: join(env.data, \"mailbox\"), tasklists: join(env.data, \"tasklists\"), teams: join(env.data, \"teams\"), wiki: join(env.data, \"wiki\"), hyperplan: join(env.data, \"hyperplan\"), dispatchGate: join(env.data, \"dispatch-gate\") }, migrationStatus: schema.migrationStatus, migrations: schema.migrations, evidence: [\"state-schema-versioning=ok\", \"state-schema-doctor=ok\", \"continuation-gate=ok\"] }\n76→}\n77→\n78→async function countSkillMarkdown(skillsDir: string): Promise<number> {\n79→  try {\n80→    const dirs = await Array.fromAsync(new Bun.Glob(\"*/SKILL.md\").scan({ cwd: skillsDir }))\n81→    return dirs.length\n82→  } catch {\n83→    return 0\n84→  }\n85→}\n86→\n87→async function teamProviderMatrix(): Promise<Array<{ provider: string; executable: string | null; available: boolean }>> {\n88→  const executables: Record<string, string | null> = { hermes: \"hermes\", claude: \"claude\", codex: \"codex\", gemini: \"gemini\", copilot: \"copilot\", zai: null, opencode: \"opencode\", grok: null, subagent: null, noop: null }\n89→  const rows: Array<{ provider: string; executable: string | null; available: boolean }> = []"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/chat_history.jsonl:35:{"type":"tool_result","tool_call_id":"call-37e12d3b-8559-4d12-b310-2a5c042f68d6-15","content":"<workspace_result workspace_path=\"/Users/ilseoblee/workspace/linalab/lfg\">\nFound at least 17 matching lines\n/Users/ilseoblee/workspace/linalab/lfg/plugins/lfg/dist/index.js\n9218:  const configProfile = resolveGrokConfigModelProfile(options);\n9220:    const policy2 = modelResolutionPolicy(agent, options.category, configProfile.profile, \"grok-config\");\n9350:function resolveGrokConfigModelProfile(options) {\n9351:  const configPath = join6(resolveLfgEnv().root, \"config\", \"config.toml\");\n9354:  const sectionNames = grokConfigModelSectionNames(options);\n9364:      provenance: { source: \"grok-config\", configPath, section: `model.${sectionName}`, baseUrl: section.base_url, envKey: section.env_key }\n9369:function grokConfigModelSectionNames(options) {\n10426:  const configPath = join17(home, \".grok\", \"config.toml\");\n10500:      \"You can further customize ~/.grok/agents/lina.md or add [subagents.roles.lina] in config.toml\"\n10506:  const configPath = join18(homeDir(input.processEnv), \".grok\", \"config.toml\");\n10521:  const configPath = join18(home, \".grok\", \"config.toml\");\n10697:  return { ok: true, status: \"ok\", defaultProvider: DEFAULT_MODEL_PROVIDER, modelRouter: { provider: DEFAULT_MODEL_PROVIDER, transport: defaultProviderTransport2(DEFAULT_MODEL_PROVIDER), defaultModel: PROVIDER_DEFAULT_MODELS2[DEFAULT_MODEL_PROVIDER], oracleGate: \"xai/grok\", reason: \"Approved multi-provider metadata is explicit; Grok Oracle review remains mandatory and native child spawning is manual-gated.\" }, grokOracle: { provider: \"xai\", model: PROVIDER_DEFAULT_MODELS2.xai, transport: defaultProviderTransport2(\"xai\"), authScheme: defaultProviderAuthScheme2(\"xai\") }, currentModel: await readModelSelection(context), grokBuildModelSwitch: { slash: \"/model <provider/model>\", cli: \"lfg models switch <provider/model>\", tmux: \"lfg grok-build model <provider/model>\" }, providers: selected, configuredProviders: configured, categoryModelProfiles: getEffectiveCategoryModelProfiles(), modelMatchingSource: OMO_MODEL_MATCHING_SOURCE, roleFitPolicies: OMO_ROLE_FIT_POLICIES, path: providersPa [... truncated (1037 chars total)]\n10870:    \"optional: add [model.openai-codex] from plugins/lfg/config/grok-config-template.toml when the host should route Codex-class external models\",\n10882:      globalConfigCaveat: \"LFG reports Grok setup steps only; it does not write ~/.grok/config.toml or run grok plugin install.\",\n\n/Users/ilseoblee/workspace/linalab/lfg/plugins/lfg/dist/index.js.map\n159:    \"import { existsSync, readFileSync } from \\\"node:fs\\\"\\nimport { join } from \\\"node:path\\\"\\nimport { resolveModel as resolveCoreModel, resolveModelWithFallback as resolveCoreModelWithFallback } from \\\"../omo-compat/model-core\\\"\\nimport { resolveLfgEnv, type LfgEnv } from \\\"../foundation/env\\\"\\nimport { loadOmoAgentRegistry, type OmoAgent } from \\\"./agent-registry\\\"\\nimport { isLfeCoreBoulderActive } from \\\"../adapters/lfe-core-hooks\\\"\\nimport { getLfeCoreModelProfileSuggestion } from \\\"../adapters/lfe-core-ulw\\\"\\nimport { OPENAI_CODEX_DEFAULT_MODEL } from \\\"./openai-codex\\\"\\nimport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n} from \\\"./provider-catalog\\\"\\n\\nexport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n}\\nexport type ReasoningLe [... truncated (1035 chars total)]\n179:    \"import { join, resolve } from \\\"node:path\\\"\\nimport { bootstrapState, ensureStateSchema, stateSchemaPath } from \\\"../foundation/state-schema\\\"\\nimport { commandEnv, directoryExists, executablePath, pathExists, readCatalogSkillCount, readJsonObject, type CommandContext, type JsonObject } from \\\"./common\\\"\\nimport { readCodexOAuthStatus } from \\\"../services/codex-oauth\\\"\\nimport { getLazyCodexCapabilities } from \\\"../services/lazy-capabilities\\\"\\n\\nexport async function doctor(context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const checks: JsonObject[] = []\\n  const add = (name: string, ok: boolean, evidence: string, required = true): void => { checks.push({ name, ok, required, evidence }) }\\n  const manifest = join(env.root, \\\".grok-plugin\\\", \\\"plugin.json\\\")\\n  add(\\\"grok_manifest\\\", await pathExists(manifest), manifest)\\n  const mcpConfig = join(env.root, \\\".mcp.json\\\")\\n  add(\\\"mcp_config\\\", await pathExis [... truncated (1035 chars total)]\n181:    \"import { access, copyFile, mkdir, readFile, readdir, writeFile } from \\\"node:fs/promises\\\"\\nimport { dirname, join } from \\\"node:path\\\"\\nimport type { LfgEnv } from \\\"../foundation/env\\\"\\n\\nexport type NativeAgentEnv = NodeJS.ProcessEnv\\n\\nexport type NativeAgentResult = Record<string, unknown>\\n\\nexport async function registerNativeAgent(input: { readonly agentId: string; readonly force: boolean; readonly env: LfgEnv; readonly processEnv: NativeAgentEnv }): Promise<NativeAgentResult> {\\n  const candidates = sourceCandidates(input.agentId, input.env, input.processEnv)\\n  const sourcePath = await firstExistingPath(candidates)\\n  if (!sourcePath) {\\n    return {\\n      ok: false,\\n      error: `Could not find source definition for agent \\\"${input.agentId}\\\"`,\\n      tried: candidates,\\n      hint: \\\"Run from within the lfg plugin or set GROK_PLUGIN_ROOT\\\",\\n    }\\n  }\\n\\n  const destDir = join(homeDir(input.processEnv), \\\".grok\\\", \\\"agents\\\")\\n  const destPath = join(destDir, `$ [... truncated (1035 chars total)]\n</workspace_result>"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/announcement_state.json:1:{"mcp_server_fingerprints":{"ast_grep":{"tool_count":2,"description_hash":12161962213042174405,"tool_names_hash":14054538023188616471},"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724},"grok_com_github":{"tool_count":46,"description_hash":12161962213042174405,"tool_names_hash":5955624062011338365},"lsp":{"tool_count":7,"description_hash":12161962213042174405,"tool_names_hash":16087898139124143397},"lfg-harness":{"tool_count":49,"description_hash":12161962213042174405,"tool_names_hash":12553283654516886887}},"announced_skill_names":["0-1-0-ff47fdd7:rules","help","0-1-0-ff47fdd7:lsp","0-1-0-ff47fdd7:init-deep","k-skill-setup","lfg:frontend-ui-ux","0-1-0-ff47fdd7:remove-ai-slops","0-1-0-ff47fdd7:ulw-loop","lfg:pre-publish-review","github-yeet","lfg:provider","mfds-food-safety","lfg:git-master","korean-law-search","cua-driver","lfg:lfg","lfg:agent-browser","lfg:cancel","lfg:start-work","implement","lfg:worker","0-1-0-ff47fdd7:refactor","0-1-0-ff47fdd7:start-work","lfg:github-triage","pptx","best-of-n","xlsx","0-1-0-ff47fdd7:comment-checker","0-1-0-ff47fdd7:frontend-ui-ux","design","review","lfg:doctor","lfg:team","lfg:setup","0-1-0-ff47fdd7:ulw-plan","execute-plan","create-skill","lfg:ralph","lfg:hyperplan","lfg:remove-deadcode","0-1-0-ff47fdd7:review-work","lfg:work-with-pr","lfg:omomomo","lfg:ai-slop-remover","lfg:dev-browser","0-1-0-ff47fdd7:debugging","lfg:get-unpublished-changes","lfg:playwright","lfg:publish","lina-decision-protocol","lfg:ulw","pr-babysit","lfg:team-mode","lfg:review-work","check-work","docx","lfg:plan","0-1-0-ff47fdd7:programming"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:1:{"timestamp":1780309299,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":5042,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-0","agentTimestampMs":1780309299460,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:3:{"timestamp":1780309310,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":8213,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-1","agentTimestampMs":1780309299466,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:9:{"timestamp":1780309384,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":23798,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-83","agentTimestampMs":1780309384107,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309380991,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:26:{"timestamp":1780309387,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":35142,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-114","agentTimestampMs":1780309387806,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309385163,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:37:{"timestamp":1780309388,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"tool_call_update","toolCallId":"call-0d26feb2-663c-468a-a265-d13c3e3ed978-3","status":"completed","content":[{"type":"content","content":{"type":"text","text":"1→# Grok Build Custom Models Configuration Template\n2→# Copy this file to config.toml and replace {{PLACEHOLDER}} values with your actual configuration\n3→\n4→[model.litellm]\n5→# Base URL for the litellm-compatible API endpoint\n6→# Example: \"https://api.litellm.ai/v1\"\n7→base_url = \"{{PLACEHOLDER}}\"\n8→\n9→# Environment variable name containing the API key\n10→# The runtime will read the actual key from this environment variable\n11→# Example: \"LITELLM_API_KEY\"\n12→env_key = \"{{PLACEHOLDER}}\"\n13→\n14→# Model identifier to use for this provider\n15→# Example: \"grok-1\" or \"claude-3-opus\"\n16→model = \"{{PLACEHOLDER}}\"\n17→\n18→[model.openai-compatible]\n19→# Base URL for an OpenAI-compatible API endpoint\n20→# Example: \"https://api.openai.com/v1\"\n21→base_url = \"{{PLACEHOLDER}}\"\n22→\n23→# Environment variable name containing the API key\n24→# Example: \"OPENAI_API_KEY\"\n25→env_key = \"{{PLACEHOLDER}}\"\n26→\n27→# Model identifier to use for this provider\n28→# Example: \"gpt-4-turbo\" or \"gpt-4\"\n29→model = \"{{PLACEHOLDER}}\"\n30→\n31→# openai-codex (gpt-5.5 / codex deep-specialist lane for main orchestrator)\n32→# Registered per Grok Build manual (11-custom-models.md) + LFP pi-ai \"openai-codex-responses\" pattern.\n33→# Use api_backend = \"responses\" for the Responses API shape (matches LFE-core / post-4471 codex expectations).\n34→# This is the primary execution provider for Sisyphus-family agents in the LazyCodex-compatible MVP.\n35→[model.openai-codex]\n36→model = \"openai-codex\"\n37→# base_url points at your codex-compatible or OpenAI-compatible endpoint that serves *-codex / gpt-5.x-codex models\n38→base_url = \"{{PLACEHOLDER}}\"\n39→env_key = \"{{PLACEHOLDER}}\"\n40→api_backend = \"responses\"\n41→auth_scheme = \"bearer\"\n42→# Optional: context_window, temperature, etc. per Grok manual\n43→\n44→# Codex OAuth option:\n45→# 1. Verify readiness first:\n46→#      lfg --json auth codex-oauth status\n47→# 2. If your Grok host should obtain auth through the Codex app login, configure the\n48→#    external provider explicitly. This is global Grok auth, not a per-model field.\n49→#\n50→# [auth]\n51→# auth_provider_command = \"bun /absolute/path/to/plugins/lfg/bin/codex-oauth-provider.ts --json-token\"\n52→# auth_provider_label = \"Codex OAuth\"\n53→\n54→[models]\n55→# Default model identifier to use when no model is explicitly specified\n56→# Should match one of the model sections above (e.g., \"grok\", \"litellm\", \"openai-compatible\")\n57→default = \"grok\"\n58→"}}],"rawOutput":{"type":"ReadFile","FileContent":{"content":"1→# Grok Build Custom Models Configuration Template\n2→# Copy this file to config.toml and replace {{PLACEHOLDER}} values with your actual configuration\n3→\n4→[model.litellm]\n5→# Base URL for the litellm-compatible API endpoint\n6→# Example: \"https://api.litellm.ai/v1\"\n7→base_url = \"{{PLACEHOLDER}}\"\n8→\n9→# Environment variable name containing the API key\n10→# The runtime will read the actual key from this environment variable\n11→# Example: \"LITELLM_API_KEY\"\n12→env_key = \"{{PLACEHOLDER}}\"\n13→\n14→# Model identifier to use for this provider\n15→# Example: \"grok-1\" or \"claude-3-opus\"\n16→model = \"{{PLACEHOLDER}}\"\n17→\n18→[model.openai-compatible]\n19→# Base URL for an OpenAI-compatible API endpoint\n20→# Example: \"https://api.openai.com/v1\"\n21→base_url = \"{{PLACEHOLDER}}\"\n22→\n23→# Environment variable name containing the API key\n24→# Example: \"OPENAI_API_KEY\"\n25→env_key = \"{{PLACEHOLDER}}\"\n26→\n27→# Model identifier to use for this provider\n28→# Example: \"gpt-4-turbo\" or \"gpt-4\"\n29→model = \"{{PLACEHOLDER}}\"\n30→\n31→# openai-codex (gpt-5.5 / codex deep-specialist lane for main orchestrator)\n32→# Registered per Grok Build manual (11-custom-models.md) + LFP pi-ai \"openai-codex-responses\" pattern.\n33→# Use api_backend = \"responses\" for the Responses API shape (matches LFE-core / post-4471 codex expectations).\n34→# This is the primary execution provider for Sisyphus-family agents in the LazyCodex-compatible MVP.\n35→[model.openai-codex]\n36→model = \"openai-codex\"\n37→# base_url points at your codex-compatible or OpenAI-compatible endpoint that serves *-codex / gpt-5.x-codex models\n38→base_url = \"{{PLACEHOLDER}}\"\n39→env_key = \"{{PLACEHOLDER}}\"\n40→api_backend = \"responses\"\n41→auth_scheme = \"bearer\"\n42→# Optional: context_window, temperature, etc. per Grok manual\n43→\n44→# Codex OAuth option:\n45→# 1. Verify readiness first:\n46→#      lfg --json auth codex-oauth status\n47→# 2. If your Grok host should obtain auth through the Codex app login, configure the\n48→#    external provider explicitly. This is global Grok auth, not a per-model field.\n49→#\n50→# [auth]\n51→# auth_provider_command = \"bun /absolute/path/to/plugins/lfg/bin/codex-oauth-provider.ts --json-token\"\n52→# auth_provider_label = \"Codex OAuth\"\n53→\n54→[models]\n55→# Default model identifier to use when no model is explicitly specified\n56→# Should match one of the model sections above (e.g., \"grok\", \"litellm\", \"openai-compatible\")\n57→default = \"grok\"\n58→","content_concise":"1→# Grok Build Custom Models Configuration Template\n2→# Copy this file to config.toml and replace {{PLACEHOLDER}} values with your actual configuration\n3→\n4→[model.litellm]\n5→# Base URL for the litellm-compatible API endpoint\n6→# Example: \"https://api.litellm.ai/v1\"\n7→base_url = \"{{PLACEHOLDER}}\"\n8→\n9→# Environment variable name containing the API key\n10→# The runtime will read the actual key from this environment variable\n11→# Example: \"LITELLM_API_KEY\"\n12→env_key = \"{{PLACEHOLDER}}\"\n13→\n14→# Model identifier to use for this provider\n15→# Example: \"grok-1\" or \"claude-3-opus\"\n16→model = \"{{PLACEHOLDER}}\"\n17→\n18→[model.openai-compatible]\n19→# Base URL for an OpenAI-compatible API endpoint\n20→# Example: \"https://api.openai.com/v1\"\n21→base_url = \"{{PLACEHOLDER}}\"\n22→\n23→# Environment variable name containing the API key\n24→# Example: \"OPENAI_API_KEY\"\n25→env_key = \"{{PLACEHOLDER}}\"\n26→\n27→# Model identifier to use for this provider\n28→# Example: \"gpt-4-turbo\" or \"gpt-4\"\n29→model = \"{{PLACEHOLDER}}\"\n30→\n31→# openai-codex (gpt-5.5 / codex deep-specialist lane for main orchestrator)\n32→# Registered per Grok Build manual (11-custom-models.md) + LFP pi-ai \"openai-codex-responses\" pattern.\n33→# Use api_backend = \"responses\" for the Responses API shape (matches LFE-core / post-4471 codex expectations).\n34→# This is the primary execution provider for Sisyphus-family agents in the LazyCodex-compatible MVP.\n35→[model.openai-codex]\n36→model = \"openai-codex\"\n37→# base_url points at your codex-compatible or OpenAI-compatible endpoint that serves *-codex / gpt-5.x-codex models\n38→base_url = \"{{PLACEHOLDER}}\"\n39→env_key = \"{{PLACEHOLDER}}\"\n40→api_backend = \"responses\"\n41→auth_scheme = \"bearer\"\n42→# Optional: context_window, temperature, etc. per Grok manual\n43→\n44→# Codex OAuth option:\n45→# 1. Verify readiness first:\n46→#      lfg --json auth codex-oauth status\n47→# 2. If your Grok host should obtain auth through the Codex app login, configure the\n48→#    external provider explicitly. This is global Grok auth, not a per-model field.\n49→#\n50→# [auth]\n51→# auth_provider_command = \"bun /absolute/path/to/plugins/lfg/bin/codex-oauth-provider.ts --json-token\"\n52→# auth_provider_label = \"Codex OAuth\"\n53→\n54→[models]\n55→# Default model identifier to use when no model is explicitly specified\n56→# Should match one of the model sections above (e.g., \"grok\", \"litellm\", \"openai-compatible\")\n57→default = \"grok\"\n58→","absolute_path":"/Users/ilseoblee/workspace/linalab/lfg/plugins/lfg/config/grok-config-template.toml","offset":null,"raw_output":"# Grok Build Custom Models Configuration Template\n# Copy this file to config.toml and replace {{PLACEHOLDER}} values with your actual configuration\n\n[model.litellm]\n# Base URL for the litellm-compatible API endpoint\n# Example: \"https://api.litellm.ai/v1\"\nbase_url = \"{{PLACEHOLDER}}\"\n\n# Environment variable name containing the API key\n# The runtime will read the actual key from this environment variable\n# Example: \"LITELLM_API_KEY\"\nenv_key = \"{{PLACEHOLDER}}\"\n\n# Model identifier to use for this provider\n# Example: \"grok-1\" or \"claude-3-opus\"\nmodel = \"{{PLACEHOLDER}}\"\n\n[model.openai-compatible]\n# Base URL for an OpenAI-compatible API endpoint\n# Example: \"https://api.openai.com/v1\"\nbase_url = \"{{PLACEHOLDER}}\"\n\n# Environment variable name containing the API key\n# Example: \"OPENAI_API_KEY\"\nenv_key = \"{{PLACEHOLDER}}\"\n\n# Model identifier to use for this provider\n# Example: \"gpt-4-turbo\" or \"gpt-4\"\nmodel = \"{{PLACEHOLDER}}\"\n\n# openai-codex (gpt-5.5 / codex deep-specialist lane for main orchestrator)\n# Registered per Grok Build manual (11-custom-models.md) + LFP pi-ai \"openai-codex-responses\" pattern.\n# Use api_backend = \"responses\" for the Responses API shape (matches LFE-core / post-4471 codex expectations).\n# This is the primary execution provider for Sisyphus-family agents in the LazyCodex-compatible MVP.\n[model.openai-codex]\nmodel = \"openai-codex\"\n# base_url points at your codex-compatible or OpenAI-compatible endpoint that serves *-codex / gpt-5.x-codex models\nbase_url = \"{{PLACEHOLDER}}\"\nenv_key = \"{{PLACEHOLDER}}\"\napi_backend = \"responses\"\nauth_scheme = \"bearer\"\n# Optional: context_window, temperature, etc. per Grok manual\n\n# Codex OAuth option:\n# 1. Verify readiness first:\n#      lfg --json auth codex-oauth status\n# 2. If your Grok host should obtain auth through the Codex app login, configure the\n#    external provider explicitly. This is global Grok auth, not a per-model field.\n#\n# [auth]\n# auth_provider_command = \"bun /absolute/path/to/plugins/lfg/bin/codex-oauth-provider.ts --json-token\"\n# auth_provider_label = \"Codex OAuth\"\n\n[models]\n# Default model identifier to use when no model is explicitly specified\n# Should match one of the model sections above (e.g., \"grok\", \"litellm\", \"openai-compatible\")\ndefault = \"grok\"\n","total_lines":58,"mtime_at_read":{"Known":1780143677053},"content_hash":760145786093249516}}},"_meta":{"totalTokens":35728,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-121","agentTimestampMs":1780309387992,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309385163,"turnStartMs":1780309380465,"updateType":"ToolCallUpdate","updateParams":{"toolCallId":"call-0d26feb2-663c-468a-a265-d13c3e3ed978-3","status":"Completed"}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:43:{"timestamp":1780309391,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":42903,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-142","agentTimestampMs":1780309391385,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309388719,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:60:{"timestamp":1780309394,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":45120,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-223","agentTimestampMs":1780309394782,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309392254,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:71:{"timestamp":1780309395,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"tool_call_update","toolCallId":"call-3e0ddbc2-725c-4e64-9e48-366536d984c7-9","status":"completed","content":[{"type":"content","content":{"type":"text","text":"found 27 matches"}}],"rawOutput":{"type":"GrepSearch","stdout":[60,119,111,114,107,115,112,97,99,101,95,114,101,115,117,108,116,32,119,111,114,107,115,112,97,99,101,95,112,97,116,104,61,34,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,34,62,10,70,111,117,110,100,32,97,116,32,108,101,97,115,116,32,50,55,32,109,97,116,99,104,105,110,103,32,108,105,110,101,115,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,101,118,105,100,101,110,99,101,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,110,97,116,105,118,101,45,111,112,101,110,97,105,45,115,100,107,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,45,105,110,115,112,101,99,116,45,97,102,116,101,114,45,101,110,97,98,108,101,46,106,115,111,110,10,49,52,49,52,58,32,32,32,32,32,32,32,32,34,112,97,116,104,34,58,32,34,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,34,10,49,52,53,48,58,32,32,32,32,34,117,115,101,114,80,97,116,104,34,58,32,34,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,34,44,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,101,118,105,100,101,110,99,101,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,110,97,116,105,118,101,45,111,112,101,110,97,105,45,115,100,107,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,45,105,110,115,112,101,99,116,46,106,115,111,110,10,49,52,52,52,58,32,32,32,32,32,32,32,32,34,112,97,116,104,34,58,32,34,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,34,10,49,52,53,48,58,32,32,32,32,34,117,115,101,114,80,97,116,104,34,58,32,34,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,34,44,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,101,118,105,100,101,110,99,101,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,110,97,116,105,118,101,45,111,112,101,110,97,105,45,115,100,107,47,116,97,115,107,45,56,45,114,101,118,105,101,119,45,100,105,102,102,46,116,120,116,10,50,52,58,32,45,32,96,111,112,101,110,97,105,45,99,111,100,101,120,96,32,105,115,32,110,111,119,32,97,32,102,105,114,115,116,45,99,108,97,115,115,32,97,112,112,114,111,118,101,100,32,112,114,111,118,105,100,101,114,32,43,32,109,111,100,101,108,32,40,119,105,116,104,32,71,114,111,107,32,109,97,110,117,97,108,45,99,111,109,112,108,105,97,110,116,32,96,91,109,111,100,101,108,46,111,112,101,110,97,105,45,99,111,100,101,120,93,96,32,116,101,109,112,108,97,116,101,32,115,101,99,116,105,111,110,32,117,115,105,110,103,32,96,97,112,105,95,98,97,99,107,101,110,100,32,61,32,34,114,101,115,112,111,110,115,101,115,34,96,41,46,10,52,48,58,64,64,32,45,53,52,44,55,32,43,53,52,44,55,32,64,64,32,71,114,111,107,45,115,112,101,99,105,102,105,99,32,112,105,101,99,101,115,32,40,84,85,73,32,105,110,116,101,103,114,97,116,105,111,110,44,32,115,112,97,119,110,32,101,110,118,101,108,111,112,101,44,32,79,114,97,99,108,101,32,103,97,116,101,44,32,99,111,110,102,105,103,46,116,111,109,108,10,52,52,51,58,32,32,32,47,47,32,91,109,111,100,101,108,46,111,112,101,110,97,105,45,99,111,100,101,120,93,32,115,101,99,116,105,111,110,32,102,114,111,109,32,116,104,101,32,99,111,110,102,105,103,32,116,101,109,112,108,97,116,101,32,40,114,101,115,112,111,110,115,101,115,32,98,97,99,107,101,110,100,32,112,101,114,32,71,114,111,107,32,109,97,110,117,97,108,32,43,32,76,70,80,32,112,97,114,105,116,121,41,46,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,101,118,105,100,101,110,99,101,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,110,97,116,105,118,101,45,111,112,101,110,97,105,45,115,100,107,47,116,97,115,107,45,56,45,102,105,110,97,108,46,116,120,116,10,50,56,54,58,226,156,147,32,103,114,111,107,45,99,111,110,102,105,103,45,116,101,109,112,108,97,116,101,46,116,111,109,108,32,62,32,115,104,111,117,108,100,32,99,111,110,116,97,105,110,32,114,101,113,117,105,114,101,100,32,102,105,101,108,100,115,32,105,110,32,91,109,111,100,101,108,46,108,105,116,101,108,108,109,93,10,50,56,56,58,226,156,147,32,103,114,111,107,45,99,111,110,102,105,103,45,116,101,109,112,108,97,116,101,46,116,111,109,108,32,62,32,115,104,111,117,108,100,32,99,111,110,116,97,105,110,32,114,101,113,117,105,114,101,100,32,102,105,101,108,100,115,32,105,110,32,91,109,111,100,101,108,46,111,112,101,110,97,105,45,99,111,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,100,105,115,116,47,105,110,100,101,120,46,106,115,10,57,51,53,49,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,54,40,114,101,115,111,108,118,101,76,102,103,69,110,118,40,41,46,114,111,111,116,44,32,34,99,111,110,102,105,103,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,49,48,52,50,54,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,49,55,40,104,111,109,101,44,32,34,46,103,114,111,107,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,49,48,53,48,48,58,32,32,32,32,32,32,34,89,111,117,32,99,97,110,32,102,117,114,116,104,101,114,32,99,117,115,116,111,109,105,122,101,32,126,47,46,103,114,111,107,47,97,103,101,110,116,115,47,108,105,110,97,46,109,100,32,111,114,32,97,100,100,32,91,115,117,98,97,103,101,110,116,115,46,114,111,108,101,115,46,108,105,110,97,93,32,105,110,32,99,111,110,102,105,103,46,116,111,109,108,34,10,49,48,53,48,54,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,49,56,40,104,111,109,101,68,105,114,40,105,110,112,117,116,46,112,114,111,99,101,115,115,69,110,118,41,44,32,34,46,103,114,111,107,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,49,48,53,50,49,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,49,56,40,104,111,109,101,44,32,34,46,103,114,111,107,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,49,48,55,52,54,58,32,32,123,32,119,111,114,107,102,108,111,119,58,32,34,103,114,111,107,95,109,97,110,117,97,108,95,103,97,116,101,115,34,44,32,115,116,97,116,117,115,58,32,34,109,97,110,117,97,108,45,103,97,116,101,100,34,44,32,115,117,114,102,97,99,101,58,32,34,104,111,115,116,45,110,97,116,105,118,101,32,71,114,111,107,32,99,104,105,108,100,32,115,112,97,119,110,105,110,103,32,97,110,100,32,108,105,118,101,32,101,120,116,101,114,110,97,108,32,109,111,100,101,108,32,101,120,101,99,117,116,105,111,110,34,44,32,101,118,105,100,101,110,99,101,58,32,34,126,47,46,103,114,111,107,47,100,111,99,115,47,117,115,101,114,45,103,117,105,100,101,47,48,57,45,112,108,117,103,105,110,115,46,109,100,59,32,126,47,46,103,114,111,107,47,100,111,99,115,47,117,115,101,114,45,103,117,105,100,101,47,49,49,45,99,117,115,116,111,109,45,109,111,100,101,108,115,46,109,100,59,32,100,111,99,115,47,83,77,79,75,69,46,109,100,32,109,97,110,117,97,108,32,103,97,116,101,32,110,111,116,101,115,34,32,125,10,49,48,56,55,48,58,32,32,32,32,34,111,112,116,105,111,110,97,108,58,32,97,100,100,32,91,109,111,100,101,108,46,111,112,101,110,97,105,45,99,111,100,101,120,93,32,102,114,111,109,32,112,108,117,103,105,110,115,47,108,102,103,47,99,111,110,102,105,103,47,103,114,111,107,45,99,111,110,102,105,103,45,116,101,109,112,108,97,116,101,46,116,111,109,108,32,119,104,101,110,32,116,104,101,32,104,111,115,116,32,115,104,111,117,108,100,32,114,111,117,116,101,32,67,111,100,101,120,45,99,108,97,115,115,32,101,120,116,101,114,110,97,108,32,109,111,100,101,108,115,34,44,10,49,48,56,56,50,58,32,32,32,32,32,32,103,108,111,98,97,108,67,111,110,102,105,103,67,97,118,101,97,116,58,32,34,76,70,71,32,114,101,112,111,114,116,115,32,71,114,111,107,32,115,101,116,117,112,32,115,116,101,112,115,32,111,110,108,121,59,32,105,116,32,100,111,101,115,32,110,111,116,32,119,114,105,116,101,32,126,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,32,111,114,32,114,117,110,32,103,114,111,107,32,112,108,117,103,105,110,32,105,110,115,116,97,108,108,46,34,44,10,49,48,56,57,48,58,32,32,32,32,32,32,32,32,111,112,116,105,111,110,97,108,77,111,100,101,108,58,32,96,91,109,111,100,101,108,46,111,112,101,110,97,105,45,99,111,100,101,120,93,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,100,105,115,116,47,105,110,100,101,120,46,106,115,46,109,97,112,10,49,53,57,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,101,120,105,115,116,115,83,121,110,99,44,32,114,101,97,100,70,105,108,101,83,121,110,99,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,102,115,92,34,92,110,105,109,112,111,114,116,32,123,32,106,111,105,110,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,114,101,115,111,108,118,101,77,111,100,101,108,32,97,115,32,114,101,115,111,108,118,101,67,111,114,101,77,111,100,101,108,44,32,114,101,115,111,108,118,101,77,111,100,101,108,87,105,116,104,70,97,108,108,98,97,99,107,32,97,115,32,114,101,115,111,108,118,101,67,111,114,101,77,111,100,101,108,87,105,116,104,70,97,108,108,98,97,99,107,32,125,32,102,114,111,109,32,92,34,46,46,47,111,109,111,45,99,111,109,112,97,116,47,109,111,100,101,108,45,99,111,114,101,92,34,92,110,105,109,112,111,114,116,32,123,32,114,101,115,111,108,118,101,76,102,103,69,110,118,44,32,116,121,112,101,32,76,102,103,69,110,118,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,101,110,118,92,34,92,110,105,109,112,111,114,116,32,123,32,108,111,97,100,79,109,111,65,103,101,110,116,82,101,103,105,115,116,114,121,44,32,116,121,112,101,32,79,109,111,65,103,101,110,116,32,125,32,102,114,111,109,32,92,34,46,47,97,103,101,110,116,45,114,101,103,105,115,116,114,121,92,34,92,110,105,109,112,111,114,116,32,123,32,105,115,76,102,101,67,111,114,101,66,111,117,108,100,101,114,65,99,116,105,118,101,32,125,32,102,114,111,109,32,92,34,46,46,47,97,100,97,112,116,101,114,115,47,108,102,101,45,99,111,114,101,45,104,111,111,107,115,92,34,92,110,105,109,112,111,114,116,32,123,32,103,101,116,76,102,101,67,111,114,101,77,111,100,101,108,80,114,111,102,105,108,101,83,117,103,103,101,115,116,105,111,110,32,125,32,102,114,111,109,32,92,34,46,46,47,97,100,97,112,116,101,114,115,47,108,102,101,45,99,111,114,101,45,117,108,119,92,34,92,110,105,109,112,111,114,116,32,123,32,79,80,69,78,65,73,95,67,79,68,69,88,95,68,69,70,65,85,76,84,95,77,79,68,69,76,32,125,32,102,114,111,109,32,92,34,46,47,111,112,101,110,97,105,45,99,111,100,101,120,92,34,92,110,105,109,112,111,114,116,32,123,92,110,32,32,65,80,80,82,79,86,69,68,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,83,44,92,110,32,32,99,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,77,79,68,69,76,95,80,82,79,86,73,68,69,82,95,65,76,73,65,83,69,83,44,92,110,32,32,80,82,79,86,73,68,69,82,95,68,69,70,65,85,76,84,95,77,79,68,69,76,83,44,92,110,32,32,116,121,112,101,32,67,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,116,121,112,101,32,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,125,32,102,114,111,109,32,92,34,46,47,112,114,111,118,105,100,101,114,45,99,97,116,97,108,111,103,92,34,92,110,92,110,101,120,112,111,114,116,32,123,92,110,32,32,65,80,80,82,79,86,69,68,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,83,44,92,110,32,32,99,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,77,79,68,69,76,95,80,82,79,86,73,68,69,82,95,65,76,73,65,83,69,83,44,92,110,32,32,80,82,79,86,73,68,69,82,95,68,69,70,65,85,76,84,95,77,79,68,69,76,83,44,92,110,32,32,116,121,112,101,32,67,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,116,121,112,101,32,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,125,92,110,101,120,112,111,114,116,32,116,121,112,101,32,82,101,97,115,111,110,105,110,103,76,101,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,55,57,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,106,111,105,110,44,32,114,101,115,111,108,118,101,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,44,32,101,110,115,117,114,101,83,116,97,116,101,83,99,104,101,109,97,44,32,115,116,97,116,101,83,99,104,101,109,97,80,97,116,104,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,115,116,97,116,101,45,115,99,104,101,109,97,92,34,92,110,105,109,112,111,114,116,32,123,32,99,111,109,109,97,110,100,69,110,118,44,32,100,105,114,101,99,116,111,114,121,69,120,105,115,116,115,44,32,101,120,101,99,117,116,97,98,108,101,80,97,116,104,44,32,112,97,116,104,69,120,105,115,116,115,44,32,114,101,97,100,67,97,116,97,108,111,103,83,107,105,108,108,67,111,117,110,116,44,32,114,101,97,100,74,115,111,110,79,98,106,101,99,116,44,32,116,121,112,101,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,44,32,116,121,112,101,32,74,115,111,110,79,98,106,101,99,116,32,125,32,102,114,111,109,32,92,34,46,47,99,111,109,109,111,110,92,34,92,110,105,109,112,111,114,116,32,123,32,114,101,97,100,67,111,100,101,120,79,65,117,116,104,83,116,97,116,117,115,32,125,32,102,114,111,109,32,92,34,46,46,47,115,101,114,118,105,99,101,115,47,99,111,100,101,120,45,111,97,117,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,103,101,116,76,97,122,121,67,111,100,101,120,67,97,112,97,98,105,108,105,116,105,101,115,32,125,32,102,114,111,109,32,92,34,46,46,47,115,101,114,118,105,99,101,115,47,108,97,122,121,45,99,97,112,97,98,105,108,105,116,105,101,115,92,34,92,110,92,110,101,120,112,111,114,116,32,97,115,121,110,99,32,102,117,110,99,116,105,111,110,32,100,111,99,116,111,114,40,99,111,110,116,101,120,116,58,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,32,61,32,123,125,41,58,32,80,114,111,109,105,115,101,60,74,115,111,110,79,98,106,101,99,116,62,32,123,92,110,32,32,99,111,110,115,116,32,101,110,118,32,61,32,99,111,109,109,97,110,100,69,110,118,40,99,111,110,116,101,120,116,41,92,110,32,32,97,119,97,105,116,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,40,101,110,118,41,92,110,32,32,99,111,110,115,116,32,99,104,101,99,107,115,58,32,74,115,111,110,79,98,106,101,99,116,91,93,32,61,32,91,93,92,110,32,32,99,111,110,115,116,32,97,100,100,32,61,32,40,110,97,109,101,58,32,115,116,114,105,110,103,44,32,111,107,58,32,98,111,111,108,101,97,110,44,32,101,118,105,100,101,110,99,101,58,32,115,116,114,105,110,103,44,32,114,101,113,117,105,114,101,100,32,61,32,116,114,117,101,41,58,32,118,111,105,100,32,61,62,32,123,32,99,104,101,99,107,115,46,112,117,115,104,40,123,32,110,97,109,101,44,32,111,107,44,32,114,101,113,117,105,114,101,100,44,32,101,118,105,100,101,110,99,101,32,125,41,32,125,92,110,32,32,99,111,110,115,116,32,109,97,110,105,102,101,115,116,32,61,32,106,111,105,110,40,101,110,118,46,114,111,111,116,44,32,92,34,46,103,114,111,107,45,112,108,117,103,105,110,92,34,44,32,92,34,112,108,117,103,105,110,46,106,115,111,110,92,34,41,92,110,32,32,97,100,100,40,92,34,103,114,111,107,95,109,97,110,105,102,101,115,116,92,34,44,32,97,119,97,105,116,32,112,97,116,104,69,120,105,115,116,115,40,109,97,110,105,102,101,115,116,41,44,32,109,97,110,105,102,101,115,116,41,92,110,32,32,99,111,110,115,116,32,109,99,112,67,111,110,102,105,103,32,61,32,106,111,105,110,40,101,110,118,46,114,111,111,116,44,32,92,34,46,109,99,112,46,106,115,111,110,92,34,41,92,110,32,32,97,100,100,40,92,34,109,99,112,95,99,111,110,102,105,103,92,34,44,32,97,119,97,105,116,32,112,97,116,104,69,120,105,115,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,56,49,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,97,99,99,101,115,115,44,32,99,111,112,121,70,105,108,101,44,32,109,107,100,105,114,44,32,114,101,97,100,70,105,108,101,44,32,114,101,97,100,100,105,114,44,32,119,114,105,116,101,70,105,108,101,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,102,115,47,112,114,111,109,105,115,101,115,92,34,92,110,105,109,112,111,114,116,32,123,32,100,105,114,110,97,109,101,44,32,106,111,105,110,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,116,121,112,101,32,123,32,76,102,103,69,110,118,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,101,110,118,92,34,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,78,97,116,105,118,101,65,103,101,110,116,69,110,118,32,61,32,78,111,100,101,74,83,46,80,114,111,99,101,115,115,69,110,118,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,78,97,116,105,118,101,65,103,101,110,116,82,101,115,117,108,116,32,61,32,82,101,99,111,114,100,60,115,116,114,105,110,103,44,32,117,110,107,110,111,119,110,62,92,110,92,110,101,120,112,111,114,116,32,97,115,121,110,99,32,102,117,110,99,116,105,111,110,32,114,101,103,105,115,116,101,114,78,97,116,105,118,101,65,103,101,110,116,40,105,110,112,117,116,58,32,123,32,114,101,97,100,111,110,108,121,32,97,103,101,110,116,73,100,58,32,115,116,114,105,110,103,59,32,114,101,97,100,111,110,108,121,32,102,111,114,99,101,58,32,98,111,111,108,101,97,110,59,32,114,101,97,100,111,110,108,121,32,101,110,118,58,32,76,102,103,69,110,118,59,32,114,101,97,100,111,110,108,121,32,112,114,111,99,101,115,115,69,110,118,58,32,78,97,116,105,118,101,65,103,101,110,116,69,110,118,32,125,41,58,32,80,114,111,109,105,115,101,60,78,97,116,105,118,101,65,103,101,110,116,82,101,115,117,108,116,62,32,123,92,110,32,32,99,111,110,115,116,32,99,97,110,100,105,100,97,116,101,115,32,61,32,115,111,117,114,99,101,67,97,110,100,105,100,97,116,101,115,40,105,110,112,117,116,46,97,103,101,110,116,73,100,44,32,105,110,112,117,116,46,101,110,118,44,32,105,110,112,117,116,46,112,114,111,99,101,115,115,69,110,118,41,92,110,32,32,99,111,110,115,116,32,115,111,117,114,99,101,80,97,116,104,32,61,32,97,119,97,105,116,32,102,105,114,115,116,69,120,105,115,116,105,110,103,80,97,116,104,40,99,97,110,100,105,100,97,116,101,115,41,92,110,32,32,105,102,32,40,33,115,111,117,114,99,101,80,97,116,104,41,32,123,92,110,32,32,32,32,114,101,116,117,114,110,32,123,92,110,32,32,32,32,32,32,111,107,58,32,102,97,108,115,101,44,92,110,32,32,32,32,32,32,101,114,114,111,114,58,32,96,67,111,117,108,100,32,110,111,116,32,102,105,110,100,32,115,111,117,114,99,101,32,100,101,102,105,110,105,116,105,111,110,32,102,111,114,32,97,103,101,110,116,32,92,34,36,123,105,110,112,117,116,46,97,103,101,110,116,73,100,125,92,34,96,44,92,110,32,32,32,32,32,32,116,114,105,101,100,58,32,99,97,110,100,105,100,97,116,101,115,44,92,110,32,32,32,32,32,32,104,105,110,116,58,32,92,34,82,117,110,32,102,114,111,109,32,119,105,116,104,105,110,32,116,104,101,32,108,102,103,32,112,108,117,103,105,110,32,111,114,32,115,101,116,32,71,82,79,75,95,80,76,85,71,73,78,95,82,79,79,84,92,34,44,92,110,32,32,32,32,125,92,110,32,32,125,92,110,92,110,32,32,99,111,110,115,116,32,100,101,115,116,68,105,114,32,61,32,106,111,105,110,40,104,111,109,101,68,105,114,40,105,110,112,117,116,46,112,114,111,99,101,115,115,69,110,118,41,44,32,92,34,46,103,114,111,107,92,34,44,32,92,34,97,103,101,110,116,115,92,34,41,92,110,32,32,99,111,110,115,116,32,100,101,115,116,80,97,116,104,32,61,32,106,111,105,110,40,100,101,115,116,68,105,114,44,32,96,36,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,56,52,58,32,32,32,32,34,101,120,112,111,114,116,32,116,121,112,101,32,76,97,122,121,99,111,100,101,120,87,111,114,107,102,108,111,119,83,116,97,116,117,115,32,61,32,92,34,105,109,112,108,101,109,101,110,116,101,100,92,34,32,124,32,92,34,100,111,99,117,109,101,110,116,101,100,45,101,113,117,105,118,97,108,101,110,116,92,34,32,124,32,92,34,109,97,110,117,97,108,45,103,97,116,101,100,92,34,32,124,32,92,34,111,117,116,45,111,102,45,115,99,111,112,101,92,34,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,76,97,122,121,99,111,100,101,120,87,111,114,107,102,108,111,119,82,111,119,32,61,32,123,92,110,32,32,114,101,97,100,111,110,108,121,32,119,111,114,107,102,108,111,119,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,115,116,97,116,117,115,58,32,76,97,122,121,99,111,100,101,120,87,111,114,107,102,108,111,119,83,116,97,116,117,115,92,110,32,32,114,101,97,100,111,110,108,121,32,115,117,114,102,97,99,101,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,101,118,105,100,101,110,99,101,58,32,115,116,114,105,110,103,92,110,125,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,76,97,122,121,99,111,100,101,120,83,116,97,116,117,115,82,101,115,117,108,116,32,61,32,123,92,110,32,32,114,101,97,100,111,110,108,121,32,111,107,58,32,116,114,117,101,92,110,32,32,114,101,97,100,111,110,108,121,32,115,116,97,116,117,115,58,32,92,34,111,107,92,34,92,110,32,32,114,101,97,100,111,110,108,121,32,99,111,109,109,97,110,100,58,32,92,34,108,97,122,121,99,111,100,101,120,32,115,116,97,116,117,115,92,34,92,110,32,32,114,101,97,100,111,110,108,121,32,115,111,117,114,99,101,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,114,111,119,115,58,32,76,97,122,121,99,111,100,101,120,87,111,114,107,102,108,111,119,82,111,119,91,93,92,110,32,32,114,101,97,100,111,110,108,121,32,115,117,109,109,97,114,121,58,32,123,92,110,32,32,32,32,114,101,97,100,111,110,108,121,32,116,111,116,97,108,58,32,110,117,109,98,101,114,92,110,32,32,32,32,114,101,97,100,111,110,108,121,32,105,109,112,108,101,109,101,110,116,101,100,58,32,110,117,109,98,101,114,92,110,32,32,32,32,114,101,97,100,111,110,108,121,32,100,111,99,117,109,101,110,116,101,100,69,113,117,105,118,97,108,101,110,116,58,32,110,117,109,98,101,114,92,110,32,32,32,32,114,101,97,100,111,110,108,121,32,109,97,110,117,97,108,71,97,116,101,100,58,32,110,117,109,98,101,114,92,110,32,32,32,32,114,101,97,100,111,110,108,121,32,111,117,116,79,102,83,99,111,112,101,58,32,110,117,109,98,101,114,92,110,32,32,125,92,110,125,92,110,92,110,99,111,110,115,116,32,76,65,90,89,67,79,68,69,120,95,82,79,87,83,32,61,32,91,92,110,32,32,123,32,119,111,114,107,102,108,111,119,58,32,92,34,105,110,115,116,97,108,108,95,98,111,111,116,115,116,114,97,112,92,34,44,32,115,116,97,116,117,115,58,32,92,34,100,111,99,117,109,101,110,116,101,100,45,101,113,117,105,118,97,108,101,110,116,92,34,44,32,115,117,114,102,97,99,101,58,32,92,34,108,102,103,32,115,101,116,117,112,32,99,104,101,99,107,47,115,104,111,119,47,105,110,115,116,97,108,108,45,112,108,97,110,32,112,108,117,115,32,71,114,111,107,32,112,108,117,103,105,110,32,109,97,110,105,102,101,115,116,92,34,44,32,101,118,105,100,101,110,99,101,58,32,92,34,112,108,117,103,105,110,115,47,108,102,103,47,98,105,110,47,108,102,103,46,116,115,32,115,101,116,117,112,59,32,112,108,117,103,105,110,115,47,108,102,103,47,46,99,108,97,117,100,101,45,112,108,117,103,105,110,47,112,108,117,103,105,110,46,106,115,111,110,59,32,100,111,99,115,47,83,77,79,75,69,46,109,100,92,34,32,125,44,92,110,32,32,123,32,119,111,114,107,102,108,111,119,58,32,92,34,112,114,111,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,56,55,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,106,111,105,110,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,115,116,97,116,101,45,115,99,104,101,109,97,92,34,92,110,105,109,112,111,114,116,32,123,32,97,115,82,101,99,111,114,100,44,32,99,111,109,109,97,110,100,69,110,118,44,32,99,111,112,121,84,114,101,101,44,32,100,101,116,101,99,116,82,101,112,111,44,32,112,97,116,104,69,120,105,115,116,115,44,32,114,101,97,100,74,115,111,110,44,32,114,101,97,100,80,114,111,118,105,100,101,114,83,116,97,116,101,44,32,115,101,116,117,112,80,97,116,104,44,32,117,116,99,78,111,119,44,32,119,114,105,116,101,74,115,111,110,44,32,116,121,112,101,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,44,32,116,121,112,101,32,74,115,111,110,79,98,106,101,99,116,32,125,32,102,114,111,109,32,92,34,46,47,99,111,109,109,111,110,92,34,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,83,101,116,117,112,79,112,116,105,111,110,115,32,61,32,123,32,112,108,117,103,105,110,68,105,114,63,58,32,115,116,114,105,110,103,59,32,100,114,121,82,117,110,63,58,32,98,111,111,108,101,97,110,32,125,92,110,101,120,112,111,114,116,32,116,121,112,101,32,83,101,116,117,112,73,110,115,116,97,108,108,80,108,97,110,79,112,116,105,111,110,115,32,61,32,123,32,109,97,114,107,101,116,112,108,97,99,101,63,58,32,115,116,114,105,110,103,32,125,92,110,92,110,101,120,112,111,114,116,32,97,115,121,110,99,32,102,117,110,99,116,105,111,110,32,115,101,116,117,112,40,111,112,116,105,111,110,115,58,32,83,101,116,117,112,79,112,116,105,111,110,115,32,61,32,123,125,44,32,99,111,110,116,101,120,116,58,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,32,61,32,123,125,41,58,32,80,114,111,109,105,115,101,60,74,115,111,110,79,98,106,101,99,116,62,32,123,92,110,32,32,99,111,110,115,116,32,101,110,118,32,61,32,99,111,109,109,97,110,100,69,110,118,40,99,111,110,116,101,120,116,41,92,110,32,32,97,119,97,105,116,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,40,101,110,118,41,92,110,32,32,99,111,110,115,116,32,100,101,115,116,32,61,32,111,112,116,105,111,110,115,46,112,108,117,103,105,110,68,105,114,32,63,63,32,112,114,111,99,101,115,115,46,101,110,118,46,76,70,71,95,80,76,85,71,73,78,95,68,69,83,84,32,63,63,32,106,111,105,110,40,112,114,111,99,101,115,115,46,101,110,118,46,72,79,77,69,32,63,63,32,92,34,92,34,44,32,92,34,46,103,114,111,107,92,34,44,32,92,34,112,108,117,103,105,110,115,92,34,44,32,92,34,108,102,103,92,34,41,92,110,32,32,99,111,110,115,116,32,100,114,121,82,117,110,32,61,32,66,111,111,108,101,97,110,40,111,112,116,105,111,110,115,46,100,114,121,82,117,110,41,92,110,32,32,99,111,110,115,116,32,112,114,111,118,105,100,101,114,83,116,97,116,101,32,61,32,97,119,97,105,116,32,114,101,97,100,80,114,111,118,105,100,101,114,83,116,97,116,101,40,101,110,118,41,92,110,32,32,105,102,32,40,33,100,114,121,82,117,110,41,32,97,119,97,105,116,32,99,111,112,121,84,114,101,101,40,101,110,118,46,114,111,111,116,44,32,100,101,115,116,41,92,110,32,32,99,111,110,115,116,32,114,101,99,111,114,100,58,32,74,115,111,110,79,98,106,101,99,116,32,61,32,123,32,111,107,58,32,116,114,117,101,44,32,100,114,121,82,117,110,44,32,105,110,115,116,97,108,108,101,100,58,32,33,100,114,121,82,117,110,44,32,112,108,117,103,105,110,58,32,123,32,115,111,117,114,99,101,58,32,101,110,118,46,114,111,111,116,44,32,100,101,115,116,44,32,101,120,105,115,116,115,58,32,97,119,97,105,116,32,112,97,116,104,69,120,105,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,97,110,115,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,110,97,116,105,118,101,45,111,112,101,110,97,105,45,115,100,107,46,109,100,10,50,50,58,45,32,68,101,102,97,117,108,116,32,97,112,112,108,105,101,100,58,32,34,110,97,116,105,118,101,32,71,114,111,107,32,66,117,105,108,100,32,115,117,112,112,111,114,116,34,32,109,101,97,110,115,32,71,114,111,107,32,112,108,117,103,105,110,47,115,107,105,108,108,47,97,103,101,110,116,47,104,111,111,107,47,77,67,80,47,99,117,115,116,111,109,45,109,111,100,101,108,32,115,117,114,102,97,99,101,115,32,97,114,101,32,100,105,115,99,111,118,101,114,97,98,108,101,32,97,110,100,32,116,101,115,116,45,98,97,99,107,101,100,59,32,114,101,97,108,32,104,111,115,116,32,99,104,105,108,100,32,115,112,97,119,110,105,110,103,32,114,101,109,97,105,110,115,32,109,97,110,117,97,108,45,103,97,116,101,100,32,117,110,108,101,115,115,32,97,32,71,114,111,107,32,116,114,97,110,115,99,114,105,112,116,32,112,114,111,118,101,115,32,105,116,46,10,50,49,57,58,32,32,45,32,96,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,46,103,114,111,107,47,100,111,99,115,47,117,115,101,114,45,103,117,105,100,101,47,49,49,45,99,117,115,116,111,109,45,109,111,100,101,108,115,46,109,100,58,53,54,96,32,45,32,71,114,111,107,32,65,80,73,32,98,97,99,107,101,110,100,32,111,112,116,105,111,110,115,46,10,50,50,48,58,32,32,45,32,96,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,46,103,114,111,107,47,100,111,99,115,47,117,115,101,114,45,103,117,105,100,101,47,49,49,45,99,117,115,116,111,109,45,109,111,100,101,108,115,46,109,100,58,54,56,96,32,45,32,71,114,111,107,32,97,117,116,104,32,115,99,104,101,109,101,32,111,112,116,105,111,110,115,46,10,50,56,54,58,32,32,42,42,87,104,97,116,32,116,111,32,100,111,42,42,58,32,85,112,100,97,116,101,32,115,101,116,117,112,47,115,116,97,116,117,115,32,100,111,99,115,32,97,110,100,32,100,101,116,101,114,109,105,110,105,115,116,105,99,32,115,101,116,117,112,32,111,117,116,112,117,116,32,115,111,32,96,108,102,103,32,115,101,116,117,112,32,99,104,101,99,107,96,32,111,114,32,96,108,102,103,32,115,101,116,117,112,32,105,110,115,116,97,108,108,45,112,108,97,110,96,32,114,101,112,111,114,116,115,32,116,104,101,32,101,120,97,99,116,32,71,114,111,107,32,115,116,101,112,115,32,102,111,114,32,76,97,122,121,67,111,100,101,120,45,99,111,109,112,97,116,105,98,108,101,32,112,108,117,103,105,110,32,117,115,101,58,32,112,108,117,103,105,110,32,105,110,115,116,97,108,108,47,116,114,117,115,116,44,32,115,107,105,108,108,32,100,105,115,99,111,118,101,114,121,44,32,77,67,80,32,100,105,115,99,111,118,101,114,121,44,32,111,112,116,105,111,110,97,108,32,96,91,97,117,116,104,93,32,97,117,116,104,95,112,114,111,118,105,100,101,114,95,99,111,109,109,97,110,100,96,44,32,111,112,116,105,111,110,97,108,32,96,91,109,111,100,101,108,46,111,112,101,110,97,105,45,99,111,100,101,120,93,96,44,32,97,110,100,32,109,97,110,117,97,108,32,71,114,111,107,32,103,97,116,101,115,46,32,73,102,32,96,115,101,116,117,112,96,32,99,111,109,109,97,110,100,32,111,117,116,112,117,116,32,105,115,32,97,108,114,101,97,100,121,32,116,104,101,32,105,110,116,101,110,100,101,100,32,112,108,97,99,101,44,32,101,120,116,101,110,100,32,105,116,59,32,111,116,104,101,114,119,105,115,101,32,97,100,100,32,97,32,96,108,97,122,121,99,111,100,101,120,96,32,115,101,99,116,105,111,110,32,116,111,32,96,100,111,99,116,111,114,96,46,10,60,47,119,111,114,107,115,112,97,99,101,95,114,101,115,117,108,116,62],"stderr":[],"exit_code":0,"match_count":27,"file_matches":[{"path":"/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/lazycodex-grok-inspect-after-enable.json","matches":[{"line_number":1414,"content":"        \"path\": \"/Users/ilseoblee/.grok/config.toml\""},{"line_number":1450,"content":"    \"userPath\": \"/Users/ilseoblee/.grok/config.toml\","}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/lazycodex-grok-inspect.json","matches":[{"line_number":1444,"content":"        \"path\": \"/Users/ilseoblee/.grok/config.toml\""},{"line_number":1450,"content":"    \"userPath\": \"/Users/ilseoblee/.grok/config.toml\","}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/task-8-review-diff.txt","matches":[{"line_number":24,"content":" - `openai-codex` is now a first-class approved provider + model (with Grok manual-compliant `[model.openai-codex]` template section using `api_backend = \"responses\"`)."},{"line_number":40,"content":"@@ -54,7 +54,7 @@ Grok-specific pieces (TUI integration, spawn envelope, Oracle gate, config.toml"},{"line_number":443,"content":"   // [model.openai-codex] section from the config template (responses backend per Grok manual + LFP parity)."}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/task-8-final.txt","matches":[{"line_number":286,"content":"✓ grok-config-template.toml > should contain required fields in [model.litellm]"},{"line_number":288,"content":"✓ grok-config-template.toml > should contain required fields in [model.openai-co"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js","matches":[{"line_number":9351,"content":"  const configPath = join6(resolveLfgEnv().root, \"config\", \"config.toml\");"},{"line_number":10426,"content":"  const configPath = join17(home, \".grok\", \"config.toml\");"},{"line_number":10500,"content":"      \"You can further customize ~/.grok/agents/lina.md or add [subagents.roles.lina] in config.toml\""},{"line_number":10506,"content":"  const configPath = join18(homeDir(input.processEnv), \".grok\", \"config.toml\");"},{"line_number":10521,"content":"  const configPath = join18(home, \".grok\", \"config.toml\");"},{"line_number":10746,"content":"  { workflow: \"grok_manual_gates\", status: \"manual-gated\", surface: \"host-native Grok child spawning and live external model execution\", evidence: \"~/.grok/docs/user-guide/09-plugins.md; ~/.grok/docs/user-guide/11-custom-models.md; docs/SMOKE.md manual gate notes\" }"},{"line_number":10870,"content":"    \"optional: add [model.openai-codex] from plugins/lfg/config/grok-config-template.toml when the host should route Codex-class external models\","},{"line_number":10882,"content":"      globalConfigCaveat: \"LFG reports Grok setup steps only; it does not write ~/.grok/config.toml or run grok plugin install.\","},{"line_number":10890,"content":"        optionalModel: `[model.openai-codex]"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js.map","matches":[{"line_number":159,"content":"    \"import { existsSync, readFileSync } from \\\"node:fs\\\"\\nimport { join } from \\\"node:path\\\"\\nimport { resolveModel as resolveCoreModel, resolveModelWithFallback as resolveCoreModelWithFallback } from \\\"../omo-compat/model-core\\\"\\nimport { resolveLfgEnv, type LfgEnv } from \\\"../foundation/env\\\"\\nimport { loadOmoAgentRegistry, type OmoAgent } from \\\"./agent-registry\\\"\\nimport { isLfeCoreBoulderActive } from \\\"../adapters/lfe-core-hooks\\\"\\nimport { getLfeCoreModelProfileSuggestion } from \\\"../adapters/lfe-core-ulw\\\"\\nimport { OPENAI_CODEX_DEFAULT_MODEL } from \\\"./openai-codex\\\"\\nimport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n} from \\\"./provider-catalog\\\"\\n\\nexport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n}\\nexport type ReasoningLevel  [... truncated (1031 chars total)]"},{"line_number":179,"content":"    \"import { join, resolve } from \\\"node:path\\\"\\nimport { bootstrapState, ensureStateSchema, stateSchemaPath } from \\\"../foundation/state-schema\\\"\\nimport { commandEnv, directoryExists, executablePath, pathExists, readCatalogSkillCount, readJsonObject, type CommandContext, type JsonObject } from \\\"./common\\\"\\nimport { readCodexOAuthStatus } from \\\"../services/codex-oauth\\\"\\nimport { getLazyCodexCapabilities } from \\\"../services/lazy-capabilities\\\"\\n\\nexport async function doctor(context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const checks: JsonObject[] = []\\n  const add = (name: string, ok: boolean, evidence: string, required = true): void => { checks.push({ name, ok, required, evidence }) }\\n  const manifest = join(env.root, \\\".grok-plugin\\\", \\\"plugin.json\\\")\\n  add(\\\"grok_manifest\\\", await pathExists(manifest), manifest)\\n  const mcpConfig = join(env.root, \\\".mcp.json\\\")\\n  add(\\\"mcp_config\\\", await pathExists(m [... truncated (1031 chars total)]"},{"line_number":181,"content":"    \"import { access, copyFile, mkdir, readFile, readdir, writeFile } from \\\"node:fs/promises\\\"\\nimport { dirname, join } from \\\"node:path\\\"\\nimport type { LfgEnv } from \\\"../foundation/env\\\"\\n\\nexport type NativeAgentEnv = NodeJS.ProcessEnv\\n\\nexport type NativeAgentResult = Record<string, unknown>\\n\\nexport async function registerNativeAgent(input: { readonly agentId: string; readonly force: boolean; readonly env: LfgEnv; readonly processEnv: NativeAgentEnv }): Promise<NativeAgentResult> {\\n  const candidates = sourceCandidates(input.agentId, input.env, input.processEnv)\\n  const sourcePath = await firstExistingPath(candidates)\\n  if (!sourcePath) {\\n    return {\\n      ok: false,\\n      error: `Could not find source definition for agent \\\"${input.agentId}\\\"`,\\n      tried: candidates,\\n      hint: \\\"Run from within the lfg plugin or set GROK_PLUGIN_ROOT\\\",\\n    }\\n  }\\n\\n  const destDir = join(homeDir(input.processEnv), \\\".grok\\\", \\\"agents\\\")\\n  const destPath = join(destDir, `${inp [... truncated (1031 chars total)]"},{"line_number":184,"content":"    \"export type LazycodexWorkflowStatus = \\\"implemented\\\" | \\\"documented-equivalent\\\" | \\\"manual-gated\\\" | \\\"out-of-scope\\\"\\n\\nexport type LazycodexWorkflowRow = {\\n  readonly workflow: string\\n  readonly status: LazycodexWorkflowStatus\\n  readonly surface: string\\n  readonly evidence: string\\n}\\n\\nexport type LazycodexStatusResult = {\\n  readonly ok: true\\n  readonly status: \\\"ok\\\"\\n  readonly command: \\\"lazycodex status\\\"\\n  readonly source: string\\n  readonly rows: LazycodexWorkflowRow[]\\n  readonly summary: {\\n    readonly total: number\\n    readonly implemented: number\\n    readonly documentedEquivalent: number\\n    readonly manualGated: number\\n    readonly outOfScope: number\\n  }\\n}\\n\\nconst LAZYCODEx_ROWS = [\\n  { workflow: \\\"install_bootstrap\\\", status: \\\"documented-equivalent\\\", surface: \\\"lfg setup check/show/install-plan plus Grok plugin manifest\\\", evidence: \\\"plugins/lfg/bin/lfg.ts setup; plugins/lfg/.claude-plugin/plugin.json; docs/SMOKE.md\\\" },\\n  { workflow: \\\"project [... truncated (1031 chars total)]"},{"line_number":187,"content":"    \"import { join } from \\\"node:path\\\"\\nimport { bootstrapState } from \\\"../foundation/state-schema\\\"\\nimport { asRecord, commandEnv, copyTree, detectRepo, pathExists, readJson, readProviderState, setupPath, utcNow, writeJson, type CommandContext, type JsonObject } from \\\"./common\\\"\\n\\nexport type SetupOptions = { pluginDir?: string; dryRun?: boolean }\\nexport type SetupInstallPlanOptions = { marketplace?: string }\\n\\nexport async function setup(options: SetupOptions = {}, context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const dest = options.pluginDir ?? process.env.LFG_PLUGIN_DEST ?? join(process.env.HOME ?? \\\"\\\", \\\".grok\\\", \\\"plugins\\\", \\\"lfg\\\")\\n  const dryRun = Boolean(options.dryRun)\\n  const providerState = await readProviderState(env)\\n  if (!dryRun) await copyTree(env.root, dest)\\n  const record: JsonObject = { ok: true, dryRun, installed: !dryRun, plugin: { source: env.root, dest, exists: await pathExists( [... truncated (1031 chars total)]"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plans/lazycodex-grokbuild-native-openai-sdk.md","matches":[{"line_number":22,"content":"- Default applied: \"native Grok Build support\" means Grok plugin/skill/agent/hook/MCP/custom-model surfaces are discoverable and test-backed; real host child spawning remains manual-gated unless a Grok transcript proves it."},{"line_number":219,"content":"  - `/Users/ilseoblee/.grok/docs/user-guide/11-custom-models.md:56` - Grok API backend options."},{"line_number":220,"content":"  - `/Users/ilseoblee/.grok/docs/user-guide/11-custom-models.md:68` - Grok auth scheme options."},{"line_number":286,"content":"  **What to do**: Update setup/status docs and deterministic setup output so `lfg setup check` or `lfg setup install-plan` reports the exact Grok steps for LazyCodex-compatible plugin use: plugin install/trust, skill discovery, MCP discovery, optional `[auth] auth_provider_command`, optional `[model.openai-codex]`, and manual Grok gates. If `setup` command output is already the intended place, extend it; otherwise add a `lazycodex` section to `doctor`."}]}]}},"_meta":{"totalTokens":45595,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-230","agentTimestampMs":1780309395002,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309392254,"turnStartMs":1780309380465,"updateType":"ToolCallUpdate","updateParams":{"toolCallId":"call-3e0ddbc2-725c-4e64-9e48-366536d984c7-9","status":"Completed"}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:75:{"timestamp":1780309395,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"tool_call_update","toolCallId":"call-3e0ddbc2-725c-4e64-9e48-366536d984c7-11","status":"completed","content":[{"type":"content","content":{"type":"text","text":"found 21 matches"}}],"rawOutput":{"type":"GrepSearch","stdout":[60,119,111,114,107,115,112,97,99,101,95,114,101,115,117,108,116,32,119,111,114,107,115,112,97,99,101,95,112,97,116,104,61,34,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,34,62,10,70,111,117,110,100,32,50,49,32,109,97,116,99,104,105,110,103,32,108,105,110,101,115,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,100,105,115,116,47,105,110,100,101,120,46,106,115,10,55,51,51,52,58,32,32,99,111,110,115,116,114,117,99,116,111,114,40,123,32,98,97,115,101,85,82,76,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,34,41,44,32,97,112,105,75,101,121,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,65,80,73,95,75,69,89,34,41,32,63,63,32,110,117,108,108,44,32,97,100,109,105,110,65,80,73,75,101,121,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,65,68,77,73,78,95,75,69,89,34,41,32,63,63,32,110,117,108,108,44,32,111,114,103,97,110,105,122,97,116,105,111,110,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,79,82,71,95,73,68,34,41,32,63,63,32,110,117,108,108,44,32,112,114,111,106,101,99,116,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,80,82,79,74,69,67,84,95,73,68,34,41,32,63,63,32,110,117,108,108,44,32,119,101,98,104,111,111,107,83,101,99,114,101,116,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,87,69,66,72,79,79,75,95,83,69,67,82,69,84,34,41,32,63,63,32,110,117,108,108,44,32,119,111,114,107,108,111,97,100,73,100,101,110,116,105,116,121,44,32,46,46,46,111,112,116,115,32,125,32,61,32,123,125,41,32,123,10,55,57,53,53,58,32,32,32,32,99,111,110,115,116,114,117,99,116,111,114,40,123,32,98,97,115,101,85,82,76,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,34,41,44,32,97,112,105,75,101,121,32,61,32,114,101,97,100,69,110,118,40,34,65,90,85,82,69,95,79,80,69,78,65,73,95,65,80,73,95,75,69,89,34,41,44,32,97,112,105,86,101,114,115,105,111,110,32,61,32,114,101,97,100,69,110,118,40,34,79,80,69,78,65,73,95,65,80,73,95,86,69,82,83,73,79,78,34,41,44,32,101,110,100,112,111,105,110,116,44,32,100,101,112,108,111,121,109,101,110,116,44,32,97,122,117,114,101,65,68,84,111,107,101,110,80,114,111,118,105,100,101,114,44,32,100,97,110,103,101,114,111,117,115,108,121,65,108,108,111,119,66,114,111,119,115,101,114,44,32,46,46,46,111,112,116,115,32,125,32,61,32,123,125,41,32,123,10,56,48,54,56,58,32,32,99,111,110,115,116,32,98,97,115,101,85,114,108,32,61,32,111,112,116,105,111,110,115,46,98,97,115,101,85,114,108,32,63,63,32,79,70,70,73,67,73,65,76,95,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,59,10,56,48,55,51,58,32,32,105,102,32,40,98,97,115,101,85,114,108,32,33,61,61,32,79,70,70,73,67,73,65,76,95,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,41,32,123,10,56,49,51,51,58,118,97,114,32,67,79,68,69,88,95,79,65,85,84,72,95,80,82,79,86,73,68,69,82,32,61,32,34,111,112,101,110,97,105,45,99,111,100,101,120,34,44,32,79,70,70,73,67,73,65,76,95,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,32,61,32,34,104,116,116,112,115,58,47,47,97,112,105,46,111,112,101,110,97,105,46,99,111,109,47,118,49,34,59,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,100,105,115,116,47,105,110,100,101,120,46,106,115,46,109,97,112,10,49,52,55,58,32,32,32,32,34,47,47,32,70,105,108,101,32,103,101,110,101,114,97,116,101,100,32,102,114,111,109,32,111,117,114,32,79,112,101,110,65,80,73,32,115,112,101,99,32,98,121,32,83,116,97,105,110,108,101,115,115,46,32,83,101,101,32,67,79,78,84,82,73,66,85,84,73,78,71,46,109,100,32,102,111,114,32,100,101,116,97,105,108,115,46,92,110,118,97,114,32,95,79,112,101,110,65,73,95,105,110,115,116,97,110,99,101,115,44,32,95,97,44,32,95,79,112,101,110,65,73,95,101,110,99,111,100,101,114,44,32,95,79,112,101,110,65,73,95,98,97,115,101,85,82,76,79,118,101,114,114,105,100,100,101,110,59,92,110,105,109,112,111,114,116,32,123,32,95,95,99,108,97,115,115,80,114,105,118,97,116,101,70,105,101,108,100,71,101,116,44,32,95,95,99,108,97,115,115,80,114,105,118,97,116,101,70,105,101,108,100,83,101,116,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,116,115,108,105,98,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,117,117,105,100,52,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,117,116,105,108,115,47,117,117,105,100,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,118,97,108,105,100,97,116,101,80,111,115,105,116,105,118,101,73,110,116,101,103,101,114,44,32,105,115,65,98,115,111,108,117,116,101,85,82,76,44,32,115,97,102,101,74,83,79,78,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,117,116,105,108,115,47,118,97,108,117,101,115,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,115,108,101,101,112,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,117,116,105,108,115,47,115,108,101,101,112,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,99,97,115,116,84,111,69,114,114,111,114,44,32,105,115,65,98,111,114,116,69,114,114,111,114,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,101,114,114,111,114,115,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,103,101,116,80,108,97,116,102,111,114,109,72,101,97,100,101,114,115,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,100,101,116,101,99,116,45,112,108,97,116,102,111,114,109,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,42,32,97,115,32,83,104,105,109,115,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,115,104,105,109,115,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,42,32,97,115,32,79,112,116,115,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,114,101,113,117,101,115,116,45,111,112,116,105,111,110,115,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,115,116,114,105,110,103,105,102,121,81,117,101,114,121,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,117,116,105,108,115,47,113,117,101,114,121,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,86,69,82,83,73,79,78,32,125,32,102,114,111,109,32,92,34,46,47,118,101,114,115,105,111,110,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,42,32,97,115,32,69,114,114,111,114,115,32,102,114,111,109,32,92,34,46,47,99,111,114,101,47,101,114,114,111,114,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,42,32,97,115,32,80,97,103,105,110,97,116,105,111,110,32,102,114,111,109,32,92,34,46,47,99,111,114,101,47,112,97,103,105,110,97,116,105,111,110,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,87,111,114,107,108,111,97,100,73,100,101,110,116,105,116,121,65,117,116,104,32,125,32,102,114,111,109,32,92,34,46,47,97,117,116,104,47,119,111,114,107,108,111,97,100,45,105,100,101,110,116,105,116,121,45,97,117,116,104,46,109,106,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,52,56,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,98,117,105,108,100,72,101,97,100,101,114,115,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,104,101,97,100,101,114,115,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,42,32,97,115,32,69,114,114,111,114,115,32,102,114,111,109,32,92,34,46,47,101,114,114,111,114,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,105,115,79,98,106,44,32,114,101,97,100,69,110,118,32,125,32,102,114,111,109,32,92,34,46,47,105,110,116,101,114,110,97,108,47,117,116,105,108,115,46,109,106,115,92,34,59,92,110,105,109,112,111,114,116,32,123,32,79,112,101,110,65,73,32,125,32,102,114,111,109,32,92,34,46,47,99,108,105,101,110,116,46,109,106,115,92,34,59,92,110,47,42,42,32,65,80,73,32,67,108,105,101,110,116,32,102,111,114,32,105,110,116,101,114,102,97,99,105,110,103,32,119,105,116,104,32,116,104,101,32,65,122,117,114,101,32,79,112,101,110,65,73,32,65,80,73,46,32,42,47,92,110,101,120,112,111,114,116,32,99,108,97,115,115,32,65,122,117,114,101,79,112,101,110,65,73,32,101,120,116,101,110,100,115,32,79,112,101,110,65,73,32,123,92,110,32,32,32,32,47,42,42,92,110,32,32,32,32,32,42,32,65,80,73,32,67,108,105,101,110,116,32,102,111,114,32,105,110,116,101,114,102,97,99,105,110,103,32,119,105,116,104,32,116,104,101,32,65,122,117,114,101,32,79,112,101,110,65,73,32,65,80,73,46,92,110,32,32,32,32,32,42,92,110,32,32,32,32,32,42,32,64,112,97,114,97,109,32,123,115,116,114,105,110,103,32,124,32,117,110,100,101,102,105,110,101,100,125,32,91,111,112,116,115,46,97,112,105,86,101,114,115,105,111,110,61,112,114,111,99,101,115,115,46,101,110,118,91,39,79,80,69,78,65,73,95,65,80,73,95,86,69,82,83,73,79,78,39,93,32,63,63,32,117,110,100,101,102,105,110,101,100,93,92,110,32,32,32,32,32,42,32,64,112,97,114,97,109,32,123,115,116,114,105,110,103,32,124,32,117,110,100,101,102,105,110,101,100,125,32,91,111,112,116,115,46,101,110,100,112,111,105,110,116,61,112,114,111,99,101,115,115,46,101,110,118,91,39,65,90,85,82,69,95,79,80,69,78,65,73,95,69,78,68,80,79,73,78,84,39,93,32,63,63,32,117,110,100,101,102,105,110,101,100,93,32,45,32,89,111,117,114,32,65,122,117,114,101,32,101,110,100,112,111,105,110,116,44,32,105,110,99,108,117,100,105,110,103,32,116,104,101,32,114,101,115,111,117,114,99,101,44,32,101,46,103,46,32,96,104,116,116,112,115,58,47,47,101,120,97,109,112,108,101,45,114,101,115,111,117,114,99,101,46,97,122,117,114,101,46,111,112,101,110,97,105,46,99,111,109,47,96,92,110,32,32,32,32,32,42,32,64,112,97,114,97,109,32,123,115,116,114,105,110,103,32,124,32,117,110,100,101,102,105,110,101,100,125,32,91,111,112,116,115,46,97,112,105,75,101,121,61,112,114,111,99,101,115,115,46,101,110,118,91,39,65,90,85,82,69,95,79,80,69,78,65,73,95,65,80,73,95,75,69,89,39,93,32,63,63,32,117,110,100,101,102,105,110,101,100,93,92,110,32,32,32,32,32,42,32,64,112,97,114,97,109,32,123,115,116,114,105,110,103,32,124,32,117,110,100,101,102,105,110,101,100,125,32,111,112,116,115,46,100,101,112,108,111,121,109,101,110,116,32,45,32,65,32,109,111,100,101,108,32,100,101,112,108,111,121,109,101,110,116,44,32,105,102,32,103,105,118,101,110,44,32,115,101,116,115,32,116,104,101,32,98,97,115,101,32,99,108,105,101,110,116,32,85,82,76,32,116,111,32,105,110,99,108,117,100,101,32,96,47,100,101,112,108,111,121,109,101,110,116,115,47,123,100,101,112,108,111,121,109,101,110,116,125,96,46,92,110,32,32,32,32,32,42,32,64,112,97,114,97,109,32,123,115,116,114,105,110,103,32,124,32,110,117,108,108,32,124,32,117,110,100,101,102,105,110,101,100,125,32,91,111,112,116,115,46,111,114,103,97,110,105,122,97,116,105,111,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,53,48,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,98,117,105,108,100,67,111,100,101,120,79,65,117,116,104,80,114,111,118,105,100,101,114,80,97,121,108,111,97,100,44,32,114,101,97,100,67,111,100,101,120,79,65,117,116,104,83,116,97,116,117,115,44,32,116,121,112,101,32,67,111,100,101,120,79,65,117,116,104,83,116,97,116,117,115,32,125,32,102,114,111,109,32,92,34,46,47,99,111,100,101,120,45,111,97,117,116,104,92,34,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,79,112,101,110,65,73,83,100,107,80,114,111,98,101,77,111,100,101,32,61,32,92,34,100,114,121,45,114,117,110,92,34,32,124,32,92,34,108,105,118,101,92,34,92,110,101,120,112,111,114,116,32,116,121,112,101,32,79,112,101,110,65,73,83,100,107,80,114,111,98,101,70,101,116,99,104,32,61,32,40,105,110,112,117,116,58,32,80,97,114,97,109,101,116,101,114,115,60,116,121,112,101,111,102,32,103,108,111,98,97,108,84,104,105,115,46,102,101,116,99,104,62,91,48,93,44,32,105,110,105,116,63,58,32,80,97,114,97,109,101,116,101,114,115,60,116,121,112,101,111,102,32,103,108,111,98,97,108,84,104,105,115,46,102,101,116,99,104,62,91,49,93,41,32,61,62,32,80,114,111,109,105,115,101,60,82,101,115,112,111,110,115,101,62,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,79,112,101,110,65,73,83,100,107,80,114,111,98,101,79,112,116,105,111,110,115,32,61,32,123,92,110,32,32,114,101,97,100,111,110,108,121,32,97,117,116,104,74,115,111,110,80,97,116,104,63,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,98,97,115,101,85,114,108,63,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,101,110,118,63,58,32,82,101,97,100,111,110,108,121,60,82,101,99,111,114,100,60,115,116,114,105,110,103,44,32,115,116,114,105,110,103,32,124,32,117,110,100,101,102,105,110,101,100,62,62,92,110,32,32,114,101,97,100,111,110,108,121,32,102,101,116,99,104,63,58,32,79,112,101,110,65,73,83,100,107,80,114,111,98,101,70,101,116,99,104,92,110,32,32,114,101,97,100,111,110,108,121,32,105,110,112,117,116,63,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,109,111,100,101,63,58,32,79,112,101,110,65,73,83,100,107,80,114,111,98,101,77,111,100,101,92,110,32,32,114,101,97,100,111,110,108,121,32,109,111,100,101,108,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,112,114,111,118,105,100,101,114,63,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,116,105,109,101,111,117,116,77,115,63,58,32,110,117,109,98,101,114,92,110,125,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,79,112,101,110,65,73,83,100,107,80,114,111,98,101,82,101,115,117,108,116,32,61,32,123,92,110,32,32,114,101,97,100,111,110,108,121,32,111,107,58,32,98,111,111,108,101,97,110,92,110,32,32,114,101,97,100,111,110,108,121,32,115,116,97,116,117,115,58,32,92,34,100,114,121,45,114,117,110,92,34,32,124,32,92,34,111,107,92,34,32,124,32,92,34,98,108,111,99,107,101,100,92,34,32,124,32,92,34,117,110,97,118,97,105,108,97,98,108,101,92,34,32,124,32,92,34,102,97,105,108,101,100,92,34,92,110,32,32,114,101,97,100,111,110,108,121,32,112,114,111,118,105,100,101,114,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,109,111,100,101,108,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,98,97,115,101,85,114,108,58,32,115,116,114,105,110,103,92,110,32,32,114,101,97,100,111,110,108,121,32,109,111,100,101,58,32,79,112,101,110,65,73,83,100,107,80,114,111,98,101,77,111,100,101,92,110,32,32,114,101,97,100,111,110,108,121,32,110,101,116,119,111,114,107,67,97,108,108,101,100,58,32,98,111,111,108,101,97,110,92,110,32,32,114,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,55,57,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,106,111,105,110,44,32,114,101,115,111,108,118,101,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,44,32,101,110,115,117,114,101,83,116,97,116,101,83,99,104,101,109,97,44,32,115,116,97,116,101,83,99,104,101,109,97,80,97,116,104,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,115,116,97,116,101,45,115,99,104,101,109,97,92,34,92,110,105,109,112,111,114,116,32,123,32,99,111,109,109,97,110,100,69,110,118,44,32,100,105,114,101,99,116,111,114,121,69,120,105,115,116,115,44,32,101,120,101,99,117,116,97,98,108,101,80,97,116,104,44,32,112,97,116,104,69,120,105,115,116,115,44,32,114,101,97,100,67,97,116,97,108,111,103,83,107,105,108,108,67,111,117,110,116,44,32,114,101,97,100,74,115,111,110,79,98,106,101,99,116,44,32,116,121,112,101,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,44,32,116,121,112,101,32,74,115,111,110,79,98,106,101,99,116,32,125,32,102,114,111,109,32,92,34,46,47,99,111,109,109,111,110,92,34,92,110,105,109,112,111,114,116,32,123,32,114,101,97,100,67,111,100,101,120,79,65,117,116,104,83,116,97,116,117,115,32,125,32,102,114,111,109,32,92,34,46,46,47,115,101,114,118,105,99,101,115,47,99,111,100,101,120,45,111,97,117,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,103,101,116,76,97,122,121,67,111,100,101,120,67,97,112,97,98,105,108,105,116,105,101,115,32,125,32,102,114,111,109,32,92,34,46,46,47,115,101,114,118,105,99,101,115,47,108,97,122,121,45,99,97,112,97,98,105,108,105,116,105,101,115,92,34,92,110,92,110,101,120,112,111,114,116,32,97,115,121,110,99,32,102,117,110,99,116,105,111,110,32,100,111,99,116,111,114,40,99,111,110,116,101,120,116,58,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,32,61,32,123,125,41,58,32,80,114,111,109,105,115,101,60,74,115,111,110,79,98,106,101,99,116,62,32,123,92,110,32,32,99,111,110,115,116,32,101,110,118,32,61,32,99,111,109,109,97,110,100,69,110,118,40,99,111,110,116,101,120,116,41,92,110,32,32,97,119,97,105,116,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,40,101,110,118,41,92,110,32,32,99,111,110,115,116,32,99,104,101,99,107,115,58,32,74,115,111,110,79,98,106,101,99,116,91,93,32,61,32,91,93,92,110,32,32,99,111,110,115,116,32,97,100,100,32,61,32,40,110,97,109,101,58,32,115,116,114,105,110,103,44,32,111,107,58,32,98,111,111,108,101,97,110,44,32,101,118,105,100,101,110,99,101,58,32,115,116,114,105,110,103,44,32,114,101,113,117,105,114,101,100,32,61,32,116,114,117,101,41,58,32,118,111,105,100,32,61,62,32,123,32,99,104,101,99,107,115,46,112,117,115,104,40,123,32,110,97,109,101,44,32,111,107,44,32,114,101,113,117,105,114,101,100,44,32,101,118,105,100,101,110,99,101,32,125,41,32,125,92,110,32,32,99,111,110,115,116,32,109,97,110,105,102,101,115,116,32,61,32,106,111,105,110,40,101,110,118,46,114,111,111,116,44,32,92,34,46,103,114,111,107,45,112,108,117,103,105,110,92,34,44,32,92,34,112,108,117,103,105,110,46,106,115,111,110,92,34,41,92,110,32,32,97,100,100,40,92,34,103,114,111,107,95,109,97,110,105,102,101,115,116,92,34,44,32,97,119,97,105,116,32,112,97,116,104,69,120,105,115,116,115,40,109,97,110,105,102,101,115,116,41,44,32,109,97,110,105,102,101,115,116,41,92,110,32,32,99,111,110,115,116,32,109,99,112,67,111,110,102,105,103,32,61,32,106,111,105,110,40,101,110,118,46,114,111,111,116,44,32,92,34,46,109,99,112,46,106,115,111,110,92,34,41,92,110,32,32,97,100,100,40,92,34,109,99,112,95,99,111,110,102,105,103,92,34,44,32,97,119,97,105,116,32,112,97,116,104,69,120,105,115,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,56,55,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,106,111,105,110,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,115,116,97,116,101,45,115,99,104,101,109,97,92,34,92,110,105,109,112,111,114,116,32,123,32,97,115,82,101,99,111,114,100,44,32,99,111,109,109,97,110,100,69,110,118,44,32,99,111,112,121,84,114,101,101,44,32,100,101,116,101,99,116,82,101,112,111,44,32,112,97,116,104,69,120,105,115,116,115,44,32,114,101,97,100,74,115,111,110,44,32,114,101,97,100,80,114,111,118,105,100,101,114,83,116,97,116,101,44,32,115,101,116,117,112,80,97,116,104,44,32,117,116,99,78,111,119,44,32,119,114,105,116,101,74,115,111,110,44,32,116,121,112,101,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,44,32,116,121,112,101,32,74,115,111,110,79,98,106,101,99,116,32,125,32,102,114,111,109,32,92,34,46,47,99,111,109,109,111,110,92,34,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,83,101,116,117,112,79,112,116,105,111,110,115,32,61,32,123,32,112,108,117,103,105,110,68,105,114,63,58,32,115,116,114,105,110,103,59,32,100,114,121,82,117,110,63,58,32,98,111,111,108,101,97,110,32,125,92,110,101,120,112,111,114,116,32,116,121,112,101,32,83,101,116,117,112,73,110,115,116,97,108,108,80,108,97,110,79,112,116,105,111,110,115,32,61,32,123,32,109,97,114,107,101,116,112,108,97,99,101,63,58,32,115,116,114,105,110,103,32,125,92,110,92,110,101,120,112,111,114,116,32,97,115,121,110,99,32,102,117,110,99,116,105,111,110,32,115,101,116,117,112,40,111,112,116,105,111,110,115,58,32,83,101,116,117,112,79,112,116,105,111,110,115,32,61,32,123,125,44,32,99,111,110,116,101,120,116,58,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,32,61,32,123,125,41,58,32,80,114,111,109,105,115,101,60,74,115,111,110,79,98,106,101,99,116,62,32,123,92,110,32,32,99,111,110,115,116,32,101,110,118,32,61,32,99,111,109,109,97,110,100,69,110,118,40,99,111,110,116,101,120,116,41,92,110,32,32,97,119,97,105,116,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,40,101,110,118,41,92,110,32,32,99,111,110,115,116,32,100,101,115,116,32,61,32,111,112,116,105,111,110,115,46,112,108,117,103,105,110,68,105,114,32,63,63,32,112,114,111,99,101,115,115,46,101,110,118,46,76,70,71,95,80,76,85,71,73,78,95,68,69,83,84,32,63,63,32,106,111,105,110,40,112,114,111,99,101,115,115,46,101,110,118,46,72,79,77,69,32,63,63,32,92,34,92,34,44,32,92,34,46,103,114,111,107,92,34,44,32,92,34,112,108,117,103,105,110,115,92,34,44,32,92,34,108,102,103,92,34,41,92,110,32,32,99,111,110,115,116,32,100,114,121,82,117,110,32,61,32,66,111,111,108,101,97,110,40,111,112,116,105,111,110,115,46,100,114,121,82,117,110,41,92,110,32,32,99,111,110,115,116,32,112,114,111,118,105,100,101,114,83,116,97,116,101,32,61,32,97,119,97,105,116,32,114,101,97,100,80,114,111,118,105,100,101,114,83,116,97,116,101,40,101,110,118,41,92,110,32,32,105,102,32,40,33,100,114,121,82,117,110,41,32,97,119,97,105,116,32,99,111,112,121,84,114,101,101,40,101,110,118,46,114,111,111,116,44,32,100,101,115,116,41,92,110,32,32,99,111,110,115,116,32,114,101,99,111,114,100,58,32,74,115,111,110,79,98,106,101,99,116,32,61,32,123,32,111,107,58,32,116,114,117,101,44,32,100,114,121,82,117,110,44,32,105,110,115,116,97,108,108,101,100,58,32,33,100,114,121,82,117,110,44,32,112,108,117,103,105,110,58,32,123,32,115,111,117,114,99,101,58,32,101,110,118,46,114,111,111,116,44,32,100,101,115,116,44,32,101,120,105,115,116,115,58,32,97,119,97,105,116,32,112,97,116,104,69,120,105,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,115,114,99,47,114,117,110,116,105,109,101,45,116,115,47,115,101,114,118,105,99,101,115,47,111,112,101,110,97,105,45,115,100,107,45,112,114,111,98,101,46,116,115,10,52,55,58,99,111,110,115,116,32,79,70,70,73,67,73,65,76,95,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,32,61,32,34,104,116,116,112,115,58,47,47,97,112,105,46,111,112,101,110,97,105,46,99,111,109,47,118,49,34,10,53,49,58,32,32,99,111,110,115,116,32,98,97,115,101,85,114,108,32,61,32,111,112,116,105,111,110,115,46,98,97,115,101,85,114,108,32,63,63,32,79,70,70,73,67,73,65,76,95,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,10,53,54,58,32,32,105,102,32,40,98,97,115,101,85,114,108,32,33,61,61,32,79,70,70,73,67,73,65,76,95,79,80,69,78,65,73,95,66,65,83,69,95,85,82,76,41,32,123,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,82,69,65,68,77,69,46,109,100,10,54,49,58,71,114,111,107,39,115,32,96,97,117,116,104,95,112,114,111,118,105,100,101,114,95,99,111,109,109,97,110,100,96,32,105,115,32,97,32,103,108,111,98,97,108,32,96,91,97,117,116,104,93,96,32,115,101,116,116,105,110,103,44,32,110,111,116,32,97,32,112,101,114,45,109,111,100,101,108,32,102,105,101,108,100,46,32,67,111,110,102,105,103,117,114,101,32,105,116,32,101,120,112,108,105,99,105,116,108,121,32,105,110,32,96,126,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,96,32,111,114,32,96,71,82,79,75,95,65,85,84,72,95,80,82,79,86,73,68,69,82,95,67,79,77,77,65,78,68,96,59,32,76,70,71,32,100,111,101,115,32,110,111,116,32,109,117,116,97,116,101,32,116,104,97,116,32,102,105,108,101,32,97,117,116,111,109,97,116,105,99,97,108,108,121,46,32,96,45,45,106,115,111,110,45,116,111,107,101,110,96,32,101,109,105,116,115,32,111,110,108,121,32,116,104,101,32,97,99,99,101,115,115,32,116,111,107,101,110,32,74,83,79,78,32,112,97,121,108,111,97,100,32,97,110,100,32,110,101,118,101,114,32,102,111,114,119,97,114,100,115,32,116,104,101,32,99,97,99,104,101,100,32,114,101,102,114,101,115,104,32,116,111,107,101,110,46,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,82,69,65,68,77,69,46,109,100,10,49,51,56,58,84,104,101,32,71,114,111,107,32,109,97,110,117,97,108,32,101,120,112,111,115,101,115,32,96,97,117,116,104,95,112,114,111,118,105,100,101,114,95,99,111,109,109,97,110,100,96,32,97,115,32,97,32,103,108,111,98,97,108,32,96,91,97,117,116,104,93,96,32,115,101,116,116,105,110,103,32,111,114,32,96,71,82,79,75,95,65,85,84,72,95,80,82,79,86,73,68,69,82,95,67,79,77,77,65,78,68,96,32,101,110,118,105,114,111,110,109,101,110,116,32,118,97,114,105,97,98,108,101,46,32,76,70,71,32,100,111,101,115,32,110,111,116,32,109,117,116,97,116,101,32,96,126,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,96,59,32,105,102,32,121,111,117,32,119,105,114,101,32,67,111,100,101,120,32,79,65,117,116,104,32,105,110,116,111,32,71,114,111,107,44,32,99,111,110,102,105,103,117,114,101,32,105,116,32,101,120,112,108,105,99,105,116,108,121,32,97,110,100,32,118,101,114,105,102,121,32,105,116,32,105,110,32,116,104,101,32,104,111,115,116,46,32,96,45,45,106,115,111,110,45,116,111,107,101,110,96,32,101,109,105,116,115,32,111,110,108,121,32,96,123,34,97,99,99,101,115,115,95,116,111,107,101,110,34,58,34,46,46,46,34,125,96,32,111,110,32,115,116,100,111,117,116,32,115,111,32,116,104,101,32,104,111,115,116,32,110,101,118,101,114,32,114,101,99,101,105,118,101,115,32,116,104,101,32,114,101,102,114,101,115,104,32,116,111,107,101,110,32,102,114,111,109,32,116,104,101,32,67,111,100,101,120,32,97,112,112,32,99,97,99,104,101,58,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,101,118,105,100,101,110,99,101,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,110,97,116,105,118,101,45,111,112,101,110,97,105,45,115,100,107,47,116,97,115,107,45,55,45,100,111,99,115,45,113,97,46,116,120,116,10,53,53,58,82,69,65,68,77,69,46,109,100,58,49,51,56,58,84,104,101,32,71,114,111,107,32,109,97,110,117,97,108,32,101,120,112,111,115,101,115,32,96,97,117,116,104,95,112,114,111,118,105,100,101,114,95,99,111,109,109,97,110,100,96,32,97,115,32,97,32,103,108,111,98,97,108,32,96,91,97,117,116,104,93,96,32,115,101,116,116,105,110,103,32,111,114,32,96,71,82,79,75,95,65,85,84,72,95,80,82,79,86,73,68,69,82,95,67,79,77,77,65,78,68,96,32,101,110,118,105,114,111,110,109,101,110,116,32,118,97,114,105,97,98,108,101,46,32,76,70,71,32,100,111,101,115,32,110,111,116,32,109,117,116,97,116,101,32,96,126,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,96,59,32,105,102,32,121,111,117,32,119,105,114,101,32,67,111,100,101,120,32,79,65,117,116,104,32,105,110,116,111,32,71,114,111,107,44,32,99,111,110,102,105,103,117,114,101,32,105,116,32,101,120,112,108,105,99,105,116,108,121,32,97,110,100,32,118,101,114,105,102,121,32,105,116,32,105,110,32,116,104,101,32,104,111,115,116,58,10,56,50,58,112,108,117,103,105,110,115,47,108,102,103,47,82,69,65,68,77,69,46,109,100,58,54,49,58,71,114,111,107,39,115,32,96,97,117,116,104,95,112,114,111,118,105,100,101,114,95,99,111,109,109,97,110,100,96,32,105,115,32,97,32,103,108,111,98,97,108,32,96,91,97,117,116,104,93,96,32,115,101,116,116,105,110,103,44,32,110,111,116,32,97,32,112,101,114,45,109,111,100,101,108,32,102,105,101,108,100,46,32,67,111,110,102,105,103,117,114,101,32,105,116,32,101,120,112,108,105,99,105,116,108,121,32,105,110,32,96,126,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,96,32,111,114,32,96,71,82,79,75,95,65,85,84,72,95,80,82,79,86,73,68,69,82,95,67,79,77,77,65,78,68,96,59,32,76,70,71,32,100,111,101,115,32,110,111,116,32,109,117,116,97,116,101,32,116,104,97,116,32,102,105,108,101,32,97,117,116,111,109,97,116,105,99,97,108,108,121,46,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,97,110,115,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,99,111,100,101,120,45,111,97,117,116,104,46,109,100,10,50,50,51,58,32,32,42,42,87,104,97,116,32,116,111,32,100,111,42,42,58,32,85,112,100,97,116,101,32,96,112,108,117,103,105,110,115,47,108,102,103,47,115,107,105,108,108,115,47,108,102,103,96,44,32,96,112,114,111,118,105,100,101,114,96,44,32,96,108,115,112,96,44,32,96,97,115,116,45,103,114,101,112,96,44,32,114,111,111,116,32,82,69,65,68,77,69,44,32,112,108,117,103,105,110,32,82,69,65,68,77,69,44,32,97,114,99,104,105,116,101,99,116,117,114,101,32,100,111,99,115,44,32,115,109,111,107,101,32,100,111,99,115,44,32,97,110,100,32,99,111,110,102,105,103,32,116,101,109,112,108,97,116,101,46,32,68,111,99,117,109,101,110,116,32,116,104,101,32,99,111,109,98,105,110,101,100,32,115,116,111,114,121,58,32,76,97,122,121,67,111,100,101,120,45,115,116,121,108,101,32,71,114,111,107,32,66,117,105,108,100,32,112,108,117,103,105,110,44,32,96,108,105,110,97,96,32,112,117,98,108,105,99,32,115,117,114,102,97,99,101,44,32,108,97,122,121,32,77,67,80,32,116,111,111,108,115,44,32,97,110,100,32,67,111,100,101,120,32,79,65,117,116,104,32,102,111,114,32,96,111,112,101,110,97,105,45,99,111,100,101,120,96,46,32,73,110,99,108,117,100,101,32,103,108,111,98,97,108,32,71,114,111,107,32,97,117,116,104,32,99,97,118,101,97,116,32,102,111,114,32,96,97,117,116,104,95,112,114,111,118,105,100,101,114,95,99,111,109,109,97,110,100,96,46,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,115,107,105,108,108,115,47,112,114,111,118,105,100,101,114,47,83,75,73,76,76,46,109,100,10,50,57,58,45,32,42,42,71,114,111,107,32,97,117,116,104,32,99,97,118,101,97,116,42,42,58,32,71,114,111,107,39,115,32,96,97,117,116,104,95,112,114,111,118,105,100,101,114,95,99,111,109,109,97,110,100,96,32,105,115,32,103,108,111,98,97,108,32,96,91,97,117,116,104,93,96,32,99,111,110,102,105,103,117,114,97,116,105,111,110,32,111,114,32,96,71,82,79,75,95,65,85,84,72,95,80,82,79,86,73,68,69,82,95,67,79,77,77,65,78,68,96,44,32,110,111,116,32,112,114,111,118,105,100,101,114,32,109,101,116,97,100,97,116,97,32,115,116,111,114,101,100,32,98,121,32,96,47,112,114,111,118,105,100,101,114,32,97,100,100,96,46,32,76,70,71,32,100,111,101,115,32,110,111,116,32,101,100,105,116,32,96,126,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,96,32,97,117,116,111,109,97,116,105,99,97,108,108,121,46,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,112,108,117,103,105,110,115,47,108,102,103,47,99,111,110,102,105,103,47,103,114,111,107,45,99,111,110,102,105,103,45,116,101,109,112,108,97,116,101,46,116,111,109,108,10,51,55,58,35,32,98,97,115,101,95,117,114,108,32,112,111,105,110,116,115,32,97,116,32,121,111,117,114,32,99,111,100,101,120,45,99,111,109,112,97,116,105,98,108,101,32,111,114,32,79,112,101,110,65,73,45,99,111,109,112,97,116,105,98,108,101,32,101,110,100,112,111,105,110,116,32,116,104,97,116,32,115,101,114,118,101,115,32,42,45,99,111,100,101,120,32,47,32,103,112,116,45,53,46,120,45,99,111,100,101,120,32,109,111,100,101,108,115,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,46,47,101,118,105,100,101,110,99,101,47,108,97,122,121,99,111,100,101,120,45,103,114,111,107,98,117,105,108,100,45,99,111,100,101,120,45,111,97,117,116,104,45,49,45,98,97,115,101,108,105,110,101,46,116,120,116,10,49,49,55,52,58,35,32,98,97,115,101,95,117,114,108,32,112,111,105,110,116,115,32,97,116,32,121,111,117,114,32,99,111,100,101,120,45,99,111,109,112,97,116,105,98,108,101,32,111,114,32,79,112,101,110,65,73,45,99,111,109,112,97,116,105,98,108,101,32,101,110,100,112,111,105,110,116,32,116,104,97,116,32,115,101,10,60,47,119,111,114,107,115,112,97,99,101,95,114,101,115,117,108,116,62],"stderr":[],"exit_code":0,"match_count":21,"file_matches":[{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js","matches":[{"line_number":7334,"content":"  constructor({ baseURL = readEnv(\"OPENAI_BASE_URL\"), apiKey = readEnv(\"OPENAI_API_KEY\") ?? null, adminAPIKey = readEnv(\"OPENAI_ADMIN_KEY\") ?? null, organization = readEnv(\"OPENAI_ORG_ID\") ?? null, project = readEnv(\"OPENAI_PROJECT_ID\") ?? null, webhookSecret = readEnv(\"OPENAI_WEBHOOK_SECRET\") ?? null, workloadIdentity, ...opts } = {}) {"},{"line_number":7955,"content":"    constructor({ baseURL = readEnv(\"OPENAI_BASE_URL\"), apiKey = readEnv(\"AZURE_OPENAI_API_KEY\"), apiVersion = readEnv(\"OPENAI_API_VERSION\"), endpoint, deployment, azureADTokenProvider, dangerouslyAllowBrowser, ...opts } = {}) {"},{"line_number":8068,"content":"  const baseUrl = options.baseUrl ?? OFFICIAL_OPENAI_BASE_URL;"},{"line_number":8073,"content":"  if (baseUrl !== OFFICIAL_OPENAI_BASE_URL) {"},{"line_number":8133,"content":"var CODEX_OAUTH_PROVIDER = \"openai-codex\", OFFICIAL_OPENAI_BASE_URL = \"https://api.openai.com/v1\";"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/dist/index.js.map","matches":[{"line_number":147,"content":"    \"// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.\\nvar _OpenAI_instances, _a, _OpenAI_encoder, _OpenAI_baseURLOverridden;\\nimport { __classPrivateFieldGet, __classPrivateFieldSet } from \\\"./internal/tslib.mjs\\\";\\nimport { uuid4 } from \\\"./internal/utils/uuid.mjs\\\";\\nimport { validatePositiveInteger, isAbsoluteURL, safeJSON } from \\\"./internal/utils/values.mjs\\\";\\nimport { sleep } from \\\"./internal/utils/sleep.mjs\\\";\\nimport { castToError, isAbortError } from \\\"./internal/errors.mjs\\\";\\nimport { getPlatformHeaders } from \\\"./internal/detect-platform.mjs\\\";\\nimport * as Shims from \\\"./internal/shims.mjs\\\";\\nimport * as Opts from \\\"./internal/request-options.mjs\\\";\\nimport { stringifyQuery } from \\\"./internal/utils/query.mjs\\\";\\nimport { VERSION } from \\\"./version.mjs\\\";\\nimport * as Errors from \\\"./core/error.mjs\\\";\\nimport * as Pagination from \\\"./core/pagination.mjs\\\";\\nimport { WorkloadIdentityAuth } from \\\"./auth/workload-identity-auth.mjs\\\"; [... truncated (1031 chars total)]"},{"line_number":148,"content":"    \"import { buildHeaders } from \\\"./internal/headers.mjs\\\";\\nimport * as Errors from \\\"./error.mjs\\\";\\nimport { isObj, readEnv } from \\\"./internal/utils.mjs\\\";\\nimport { OpenAI } from \\\"./client.mjs\\\";\\n/** API Client for interfacing with the Azure OpenAI API. */\\nexport class AzureOpenAI extends OpenAI {\\n    /**\\n     * API Client for interfacing with the Azure OpenAI API.\\n     *\\n     * @param {string | undefined} [opts.apiVersion=process.env['OPENAI_API_VERSION'] ?? undefined]\\n     * @param {string | undefined} [opts.endpoint=process.env['AZURE_OPENAI_ENDPOINT'] ?? undefined] - Your Azure endpoint, including the resource, e.g. `https://example-resource.azure.openai.com/`\\n     * @param {string | undefined} [opts.apiKey=process.env['AZURE_OPENAI_API_KEY'] ?? undefined]\\n     * @param {string | undefined} opts.deployment - A model deployment, if given, sets the base client URL to include `/deployments/{deployment}`.\\n     * @param {string | null | undefined} [opts.organization=pr [... truncated (1031 chars total)]"},{"line_number":150,"content":"    \"import { buildCodexOAuthProviderPayload, readCodexOAuthStatus, type CodexOAuthStatus } from \\\"./codex-oauth\\\"\\n\\nexport type OpenAISdkProbeMode = \\\"dry-run\\\" | \\\"live\\\"\\nexport type OpenAISdkProbeFetch = (input: Parameters<typeof globalThis.fetch>[0], init?: Parameters<typeof globalThis.fetch>[1]) => Promise<Response>\\n\\nexport type OpenAISdkProbeOptions = {\\n  readonly authJsonPath?: string\\n  readonly baseUrl?: string\\n  readonly env?: Readonly<Record<string, string | undefined>>\\n  readonly fetch?: OpenAISdkProbeFetch\\n  readonly input?: string\\n  readonly mode?: OpenAISdkProbeMode\\n  readonly model: string\\n  readonly provider?: string\\n  readonly timeoutMs?: number\\n}\\n\\nexport type OpenAISdkProbeResult = {\\n  readonly ok: boolean\\n  readonly status: \\\"dry-run\\\" | \\\"ok\\\" | \\\"blocked\\\" | \\\"unavailable\\\" | \\\"failed\\\"\\n  readonly provider: string\\n  readonly model: string\\n  readonly baseUrl: string\\n  readonly mode: OpenAISdkProbeMode\\n  readonly networkCalled: boolean\\n  reado [... truncated (1031 chars total)]"},{"line_number":179,"content":"    \"import { join, resolve } from \\\"node:path\\\"\\nimport { bootstrapState, ensureStateSchema, stateSchemaPath } from \\\"../foundation/state-schema\\\"\\nimport { commandEnv, directoryExists, executablePath, pathExists, readCatalogSkillCount, readJsonObject, type CommandContext, type JsonObject } from \\\"./common\\\"\\nimport { readCodexOAuthStatus } from \\\"../services/codex-oauth\\\"\\nimport { getLazyCodexCapabilities } from \\\"../services/lazy-capabilities\\\"\\n\\nexport async function doctor(context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const checks: JsonObject[] = []\\n  const add = (name: string, ok: boolean, evidence: string, required = true): void => { checks.push({ name, ok, required, evidence }) }\\n  const manifest = join(env.root, \\\".grok-plugin\\\", \\\"plugin.json\\\")\\n  add(\\\"grok_manifest\\\", await pathExists(manifest), manifest)\\n  const mcpConfig = join(env.root, \\\".mcp.json\\\")\\n  add(\\\"mcp_config\\\", await pathExists(m [... truncated (1031 chars total)]"},{"line_number":187,"content":"    \"import { join } from \\\"node:path\\\"\\nimport { bootstrapState } from \\\"../foundation/state-schema\\\"\\nimport { asRecord, commandEnv, copyTree, detectRepo, pathExists, readJson, readProviderState, setupPath, utcNow, writeJson, type CommandContext, type JsonObject } from \\\"./common\\\"\\n\\nexport type SetupOptions = { pluginDir?: string; dryRun?: boolean }\\nexport type SetupInstallPlanOptions = { marketplace?: string }\\n\\nexport async function setup(options: SetupOptions = {}, context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const dest = options.pluginDir ?? process.env.LFG_PLUGIN_DEST ?? join(process.env.HOME ?? \\\"\\\", \\\".grok\\\", \\\"plugins\\\", \\\"lfg\\\")\\n  const dryRun = Boolean(options.dryRun)\\n  const providerState = await readProviderState(env)\\n  if (!dryRun) await copyTree(env.root, dest)\\n  const record: JsonObject = { ok: true, dryRun, installed: !dryRun, plugin: { source: env.root, dest, exists: await pathExists( [... truncated (1031 chars total)]"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/src/runtime-ts/services/openai-sdk-probe.ts","matches":[{"line_number":47,"content":"const OFFICIAL_OPENAI_BASE_URL = \"https://api.openai.com/v1\""},{"line_number":51,"content":"  const baseUrl = options.baseUrl ?? OFFICIAL_OPENAI_BASE_URL"},{"line_number":56,"content":"  if (baseUrl !== OFFICIAL_OPENAI_BASE_URL) {"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/README.md","matches":[{"line_number":61,"content":"Grok's `auth_provider_command` is a global `[auth]` setting, not a per-model field. Configure it explicitly in `~/.grok/config.toml` or `GROK_AUTH_PROVIDER_COMMAND`; LFG does not mutate that file automatically. `--json-token` emits only the access token JSON payload and never forwards the cached refresh token."}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./README.md","matches":[{"line_number":138,"content":"The Grok manual exposes `auth_provider_command` as a global `[auth]` setting or `GROK_AUTH_PROVIDER_COMMAND` environment variable. LFG does not mutate `~/.grok/config.toml`; if you wire Codex OAuth into Grok, configure it explicitly and verify it in the host. `--json-token` emits only `{\"access_token\":\"...\"}` on stdout so the host never receives the refresh token from the Codex app cache:"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-native-openai-sdk/task-7-docs-qa.txt","matches":[{"line_number":55,"content":"README.md:138:The Grok manual exposes `auth_provider_command` as a global `[auth]` setting or `GROK_AUTH_PROVIDER_COMMAND` environment variable. LFG does not mutate `~/.grok/config.toml`; if you wire Codex OAuth into Grok, configure it explicitly and verify it in the host:"},{"line_number":82,"content":"plugins/lfg/README.md:61:Grok's `auth_provider_command` is a global `[auth]` setting, not a per-model field. Configure it explicitly in `~/.grok/config.toml` or `GROK_AUTH_PROVIDER_COMMAND`; LFG does not mutate that file automatically."}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plans/lazycodex-grokbuild-codex-oauth.md","matches":[{"line_number":223,"content":"  **What to do**: Update `plugins/lfg/skills/lfg`, `provider`, `lsp`, `ast-grep`, root README, plugin README, architecture docs, smoke docs, and config template. Document the combined story: LazyCodex-style Grok Build plugin, `lina` public surface, lazy MCP tools, and Codex OAuth for `openai-codex`. Include global Grok auth caveat for `auth_provider_command`."}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/skills/provider/SKILL.md","matches":[{"line_number":29,"content":"- **Grok auth caveat**: Grok's `auth_provider_command` is global `[auth]` configuration or `GROK_AUTH_PROVIDER_COMMAND`, not provider metadata stored by `/provider add`. LFG does not edit `~/.grok/config.toml` automatically."}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./plugins/lfg/config/grok-config-template.toml","matches":[{"line_number":37,"content":"# base_url points at your codex-compatible or OpenAI-compatible endpoint that serves *-codex / gpt-5.x-codex models"}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/./evidence/lazycodex-grokbuild-codex-oauth-1-baseline.txt","matches":[{"line_number":1174,"content":"# base_url points at your codex-compatible or OpenAI-compatible endpoint that se"}]}]}},"_meta":{"totalTokens":47802,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-232","agentTimestampMs":1780309395097,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309392254,"turnStartMs":1780309380465,"updateType":"ToolCallUpdate","updateParams":{"toolCallId":"call-3e0ddbc2-725c-4e64-9e48-366536d984c7-11","status":"Completed"}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:77:{"timestamp":1780309399,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":50239,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-282","agentTimestampMs":1780309399015,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309395690,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:89:{"timestamp":1780309400,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":51638,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-318","agentTimestampMs":1780309400485,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309399739,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:97:{"timestamp":1780309400,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"tool_call_update","toolCallId":"call-37e12d3b-8559-4d12-b310-2a5c042f68d6-14","status":"completed","content":[{"type":"content","content":{"type":"text","text":"40→  add(\"default_launcher\", true, `lfg (effective=${env.launcher})`, false)\n41→  const schema = await ensureStateSchema(env)\n42→  add(\"state_schema\", schema.version === 2 && await pathExists(stateSchemaPath(env)), `${stateSchemaPath(env)} version=${schema.version}`, true)\n43→  const tsSource = join(env.root, \"src\", \"runtime-ts\", \"index.ts\")\n44→  add(\"typescript_runtime\", await pathExists(tsSource), `${tsSource} test=bun test plugins/lfg/src/runtime-ts/index.test.ts`, true)\n45→  const providers = await teamProviderMatrix()\n46→  const available = providers.filter((provider) => provider.available).map((provider) => provider.provider)\n47→  add(\"team_provider_commands\", true, `available=${available.join(\",\")} providers=${providers.map((provider) => provider.provider).join(\",\")}`, false)\n48→  const bridge = await hookBridgeStatus(env.root)\n49→  add(\"global_hook_bridge\", Boolean(bridge.ok), `installed=${String(bridge.installed)} valid=${String(bridge.valid)} config=${String(bridge.config)}`, false)\n50→  const codexOAuth = await readCodexOAuthStatus({ env: context.processEnv ?? process.env })\n51→  add(\"codex_oauth\", codexOAuth.ok, `status=${codexOAuth.status} auth=${codexOAuth.authPath} has_refresh=${String(codexOAuth.hasRefreshToken)}`, false)\n52→\n53→  // Lina surface agent as default check (Lina mode override)\n54→  const home = process.env.HOME || process.env.USERPROFILE || \"\"\n55→  const configPath = join(home, \".grok\", \"config.toml\")\n56→  const agentsDir = join(home, \".grok\", \"agents\")\n57→  const hasLinaDef = await pathExists(join(agentsDir, \"lina.md\"))\n58→  let linaIsDefault = false\n59→  try {\n60→    const fs = await import(\"node:fs/promises\")\n61→    const cfg = await fs.readFile(configPath, \"utf8\")\n62→    linaIsDefault = /default\\s*=\\s*\"lina\"/.test(cfg)\n63→  } catch {}\n64→  const linaModeOk = hasLinaDef && linaIsDefault\n65→  add(\"lina_surface_default\", linaModeOk, `lina.md=${hasLinaDef} config_default=lina=${linaIsDefault} (run 'lfg agents register-native lina && lfg agents set-default lina' to activate full Lina mode)`, false)\n66→\n67→  const failedRequired = checks.filter((check) => check.required === true && check.ok !== true)\n68→  const warnings = checks.filter((check) => check.required !== true && check.ok !== true)\n69→  return { ok: failedRequired.length === 0, status: failedRequired.length === 0 ? \"pass\" : \"fail\", pluginRoot: env.root, pluginData: env.data, checks, capabilities: getLazyCodexCapabilities(), failedRequired, warnings }\n70→}\n71→\n72→export async function doctorStateSchemaCheck(context: CommandContext = {}): Promise<JsonObject> {\n73→  const env = commandEnv(context)\n74→  const schema = await bootstrapState(env)\n75→  return { ok: true, status: \"pass\", operation: \"doctor_state_schema_check\", schema, stateRoots: { state: env.stateDir, runs: env.runsDir, plans: env.plansDir, boulder: join(env.data, \"boulder\"), notepads: join(env.data, \"notepads\"), mailbox: join(env.data, \"mailbox\"), tasklists: join(env.data, \"tasklists\"), teams: join(env.data, \"teams\"), wiki: join(env.data, \"wiki\"), hyperplan: join(env.data, \"hyperplan\"), dispatchGate: join(env.data, \"dispatch-gate\") }, migrationStatus: schema.migrationStatus, migrations: schema.migrations, evidence: [\"state-schema-versioning=ok\", \"state-schema-doctor=ok\", \"continuation-gate=ok\"] }\n76→}\n77→\n78→async function countSkillMarkdown(skillsDir: string): Promise<number> {\n79→  try {\n80→    const dirs = await Array.fromAsync(new Bun.Glob(\"*/SKILL.md\").scan({ cwd: skillsDir }))\n81→    return dirs.length\n82→  } catch {\n83→    return 0\n84→  }\n85→}\n86→\n87→async function teamProviderMatrix(): Promise<Array<{ provider: string; executable: string | null; available: boolean }>> {\n88→  const executables: Record<string, string | null> = { hermes: \"hermes\", claude: \"claude\", codex: \"codex\", gemini: \"gemini\", copilot: \"copilot\", zai: null, opencode: \"opencode\", grok: null, subagent: null, noop: null }\n89→  const rows: Array<{ provider: string; executable: string | null; available: boolean }> = []"}}],"rawOutput":{"type":"ReadFile","FileContent":{"content":"40→  add(\"default_launcher\", true, `lfg (effective=${env.launcher})`, false)\n41→  const schema = await ensureStateSchema(env)\n42→  add(\"state_schema\", schema.version === 2 && await pathExists(stateSchemaPath(env)), `${stateSchemaPath(env)} version=${schema.version}`, true)\n43→  const tsSource = join(env.root, \"src\", \"runtime-ts\", \"index.ts\")\n44→  add(\"typescript_runtime\", await pathExists(tsSource), `${tsSource} test=bun test plugins/lfg/src/runtime-ts/index.test.ts`, true)\n45→  const providers = await teamProviderMatrix()\n46→  const available = providers.filter((provider) => provider.available).map((provider) => provider.provider)\n47→  add(\"team_provider_commands\", true, `available=${available.join(\",\")} providers=${providers.map((provider) => provider.provider).join(\",\")}`, false)\n48→  const bridge = await hookBridgeStatus(env.root)\n49→  add(\"global_hook_bridge\", Boolean(bridge.ok), `installed=${String(bridge.installed)} valid=${String(bridge.valid)} config=${String(bridge.config)}`, false)\n50→  const codexOAuth = await readCodexOAuthStatus({ env: context.processEnv ?? process.env })\n51→  add(\"codex_oauth\", codexOAuth.ok, `status=${codexOAuth.status} auth=${codexOAuth.authPath} has_refresh=${String(codexOAuth.hasRefreshToken)}`, false)\n52→\n53→  // Lina surface agent as default check (Lina mode override)\n54→  const home = process.env.HOME || process.env.USERPROFILE || \"\"\n55→  const configPath = join(home, \".grok\", \"config.toml\")\n56→  const agentsDir = join(home, \".grok\", \"agents\")\n57→  const hasLinaDef = await pathExists(join(agentsDir, \"lina.md\"))\n58→  let linaIsDefault = false\n59→  try {\n60→    const fs = await import(\"node:fs/promises\")\n61→    const cfg = await fs.readFile(configPath, \"utf8\")\n62→    linaIsDefault = /default\\s*=\\s*\"lina\"/.test(cfg)\n63→  } catch {}\n64→  const linaModeOk = hasLinaDef && linaIsDefault\n65→  add(\"lina_surface_default\", linaModeOk, `lina.md=${hasLinaDef} config_default=lina=${linaIsDefault} (run 'lfg agents register-native lina && lfg agents set-default lina' to activate full Lina mode)`, false)\n66→\n67→  const failedRequired = checks.filter((check) => check.required === true && check.ok !== true)\n68→  const warnings = checks.filter((check) => check.required !== true && check.ok !== true)\n69→  return { ok: failedRequired.length === 0, status: failedRequired.length === 0 ? \"pass\" : \"fail\", pluginRoot: env.root, pluginData: env.data, checks, capabilities: getLazyCodexCapabilities(), failedRequired, warnings }\n70→}\n71→\n72→export async function doctorStateSchemaCheck(context: CommandContext = {}): Promise<JsonObject> {\n73→  const env = commandEnv(context)\n74→  const schema = await bootstrapState(env)\n75→  return { ok: true, status: \"pass\", operation: \"doctor_state_schema_check\", schema, stateRoots: { state: env.stateDir, runs: env.runsDir, plans: env.plansDir, boulder: join(env.data, \"boulder\"), notepads: join(env.data, \"notepads\"), mailbox: join(env.data, \"mailbox\"), tasklists: join(env.data, \"tasklists\"), teams: join(env.data, \"teams\"), wiki: join(env.data, \"wiki\"), hyperplan: join(env.data, \"hyperplan\"), dispatchGate: join(env.data, \"dispatch-gate\") }, migrationStatus: schema.migrationStatus, migrations: schema.migrations, evidence: [\"state-schema-versioning=ok\", \"state-schema-doctor=ok\", \"continuation-gate=ok\"] }\n76→}\n77→\n78→async function countSkillMarkdown(skillsDir: string): Promise<number> {\n79→  try {\n80→    const dirs = await Array.fromAsync(new Bun.Glob(\"*/SKILL.md\").scan({ cwd: skillsDir }))\n81→    return dirs.length\n82→  } catch {\n83→    return 0\n84→  }\n85→}\n86→\n87→async function teamProviderMatrix(): Promise<Array<{ provider: string; executable: string | null; available: boolean }>> {\n88→  const executables: Record<string, string | null> = { hermes: \"hermes\", claude: \"claude\", codex: \"codex\", gemini: \"gemini\", copilot: \"copilot\", zai: null, opencode: \"opencode\", grok: null, subagent: null, noop: null }\n89→  const rows: Array<{ provider: string; executable: string | null; available: boolean }> = []","content_concise":"40→  add(\"default_launcher\", true, `lfg (effective=${env.launcher})`, false)\n41→  const schema = await ensureStateSchema(env)\n42→  add(\"state_schema\", schema.version === 2 && await pathExists(stateSchemaPath(env)), `${stateSchemaPath(env)} version=${schema.version}`, true)\n43→  const tsSource = join(env.root, \"src\", \"runtime-ts\", \"index.ts\")\n44→  add(\"typescript_runtime\", await pathExists(tsSource), `${tsSource} test=bun test plugins/lfg/src/runtime-ts/index.test.ts`, true)\n45→  const providers = await teamProviderMatrix()\n46→  const available = providers.filter((provider) => provider.available).map((provider) => provider.provider)\n47→  add(\"team_provider_commands\", true, `available=${available.join(\",\")} providers=${providers.map((provider) => provider.provider).join(\",\")}`, false)\n48→  const bridge = await hookBridgeStatus(env.root)\n49→  add(\"global_hook_bridge\", Boolean(bridge.ok), `installed=${String(bridge.installed)} valid=${String(bridge.valid)} config=${String(bridge.config)}`, false)\n50→  const codexOAuth = await readCodexOAuthStatus({ env: context.processEnv ?? process.env })\n51→  add(\"codex_oauth\", codexOAuth.ok, `status=${codexOAuth.status} auth=${codexOAuth.authPath} has_refresh=${String(codexOAuth.hasRefreshToken)}`, false)\n52→\n53→  // Lina surface agent as default check (Lina mode override)\n54→  const home = process.env.HOME || process.env.USERPROFILE || \"\"\n55→  const configPath = join(home, \".grok\", \"config.toml\")\n56→  const agentsDir = join(home, \".grok\", \"agents\")\n57→  const hasLinaDef = await pathExists(join(agentsDir, \"lina.md\"))\n58→  let linaIsDefault = false\n59→  try {\n60→    const fs = await import(\"node:fs/promises\")\n61→    const cfg = await fs.readFile(configPath, \"utf8\")\n62→    linaIsDefault = /default\\s*=\\s*\"lina\"/.test(cfg)\n63→  } catch {}\n64→  const linaModeOk = hasLinaDef && linaIsDefault\n65→  add(\"lina_surface_default\", linaModeOk, `lina.md=${hasLinaDef} config_default=lina=${linaIsDefault} (run 'lfg agents register-native lina && lfg agents set-default lina' to activate full Lina mode)`, false)\n66→\n67→  const failedRequired = checks.filter((check) => check.required === true && check.ok !== true)\n68→  const warnings = checks.filter((check) => check.required !== true && check.ok !== true)\n69→  return { ok: failedRequired.length === 0, status: failedRequired.length === 0 ? \"pass\" : \"fail\", pluginRoot: env.root, pluginData: env.data, checks, capabilities: getLazyCodexCapabilities(), failedRequired, warnings }\n70→}\n71→\n72→export async function doctorStateSchemaCheck(context: CommandContext = {}): Promise<JsonObject> {\n73→  const env = commandEnv(context)\n74→  const schema = await bootstrapState(env)\n75→  return { ok: true, status: \"pass\", operation: \"doctor_state_schema_check\", schema, stateRoots: { state: env.stateDir, runs: env.runsDir, plans: env.plansDir, boulder: join(env.data, \"boulder\"), notepads: join(env.data, \"notepads\"), mailbox: join(env.data, \"mailbox\"), tasklists: join(env.data, \"tasklists\"), teams: join(env.data, \"teams\"), wiki: join(env.data, \"wiki\"), hyperplan: join(env.data, \"hyperplan\"), dispatchGate: join(env.data, \"dispatch-gate\") }, migrationStatus: schema.migrationStatus, migrations: schema.migrations, evidence: [\"state-schema-versioning=ok\", \"state-schema-doctor=ok\", \"continuation-gate=ok\"] }\n76→}\n77→\n78→async function countSkillMarkdown(skillsDir: string): Promise<number> {\n79→  try {\n80→    const dirs = await Array.fromAsync(new Bun.Glob(\"*/SKILL.md\").scan({ cwd: skillsDir }))\n81→    return dirs.length\n82→  } catch {\n83→    return 0\n84→  }\n85→}\n86→\n87→async function teamProviderMatrix(): Promise<Array<{ provider: string; executable: string | null; available: boolean }>> {\n88→  const executables: Record<string, string | null> = { hermes: \"hermes\", claude: \"claude\", codex: \"codex\", gemini: \"gemini\", copilot: \"copilot\", zai: null, opencode: \"opencode\", grok: null, subagent: null, noop: null }\n89→  const rows: Array<{ provider: string; executable: string | null; available: boolean }> = []","absolute_path":"/Users/ilseoblee/workspace/linalab/lfg/plugins/lfg/src/runtime-ts/commands/doctor.ts","offset":40,"limit":50,"raw_output":"  add(\"default_launcher\", true, `lfg (effective=${env.launcher})`, false)\n  const schema = await ensureStateSchema(env)\n  add(\"state_schema\", schema.version === 2 && await pathExists(stateSchemaPath(env)), `${stateSchemaPath(env)} version=${schema.version}`, true)\n  const tsSource = join(env.root, \"src\", \"runtime-ts\", \"index.ts\")\n  add(\"typescript_runtime\", await pathExists(tsSource), `${tsSource} test=bun test plugins/lfg/src/runtime-ts/index.test.ts`, true)\n  const providers = await teamProviderMatrix()\n  const available = providers.filter((provider) => provider.available).map((provider) => provider.provider)\n  add(\"team_provider_commands\", true, `available=${available.join(\",\")} providers=${providers.map((provider) => provider.provider).join(\",\")}`, false)\n  const bridge = await hookBridgeStatus(env.root)\n  add(\"global_hook_bridge\", Boolean(bridge.ok), `installed=${String(bridge.installed)} valid=${String(bridge.valid)} config=${String(bridge.config)}`, false)\n  const codexOAuth = await readCodexOAuthStatus({ env: context.processEnv ?? process.env })\n  add(\"codex_oauth\", codexOAuth.ok, `status=${codexOAuth.status} auth=${codexOAuth.authPath} has_refresh=${String(codexOAuth.hasRefreshToken)}`, false)\n\n  // Lina surface agent as default check (Lina mode override)\n  const home = process.env.HOME || process.env.USERPROFILE || \"\"\n  const configPath = join(home, \".grok\", \"config.toml\")\n  const agentsDir = join(home, \".grok\", \"agents\")\n  const hasLinaDef = await pathExists(join(agentsDir, \"lina.md\"))\n  let linaIsDefault = false\n  try {\n    const fs = await import(\"node:fs/promises\")\n    const cfg = await fs.readFile(configPath, \"utf8\")\n    linaIsDefault = /default\\s*=\\s*\"lina\"/.test(cfg)\n  } catch {}\n  const linaModeOk = hasLinaDef && linaIsDefault\n  add(\"lina_surface_default\", linaModeOk, `lina.md=${hasLinaDef} config_default=lina=${linaIsDefault} (run 'lfg agents register-native lina && lfg agents set-default lina' to activate full Lina mode)`, false)\n\n  const failedRequired = checks.filter((check) => check.required === true && check.ok !== true)\n  const warnings = checks.filter((check) => check.required !== true && check.ok !== true)\n  return { ok: failedRequired.length === 0, status: failedRequired.length === 0 ? \"pass\" : \"fail\", pluginRoot: env.root, pluginData: env.data, checks, capabilities: getLazyCodexCapabilities(), failedRequired, warnings }\n}\n\nexport async function doctorStateSchemaCheck(context: CommandContext = {}): Promise<JsonObject> {\n  const env = commandEnv(context)\n  const schema = await bootstrapState(env)\n  return { ok: true, status: \"pass\", operation: \"doctor_state_schema_check\", schema, stateRoots: { state: env.stateDir, runs: env.runsDir, plans: env.plansDir, boulder: join(env.data, \"boulder\"), notepads: join(env.data, \"notepads\"), mailbox: join(env.data, \"mailbox\"), tasklists: join(env.data, \"tasklists\"), teams: join(env.data, \"teams\"), wiki: join(env.data, \"wiki\"), hyperplan: join(env.data, \"hyperplan\"), dispatchGate: join(env.data, \"dispatch-gate\") }, migrationStatus: schema.migrationStatus, migrations: schema.migrations, evidence: [\"state-schema-versioning=ok\", \"state-schema-doctor=ok\", \"continuation-gate=ok\"] }\n}\n\nasync function countSkillMarkdown(skillsDir: string): Promise<number> {\n  try {\n    const dirs = await Array.fromAsync(new Bun.Glob(\"*/SKILL.md\").scan({ cwd: skillsDir }))\n    return dirs.length\n  } catch {\n    return 0\n  }\n}\n\nasync function teamProviderMatrix(): Promise<Array<{ provider: string; executable: string | null; available: boolean }>> {\n  const executables: Record<string, string | null> = { hermes: \"hermes\", claude: \"claude\", codex: \"codex\", gemini: \"gemini\", copilot: \"copilot\", zai: null, opencode: \"opencode\", grok: null, subagent: null, noop: null }\n  const rows: Array<{ provider: string; executable: string | null; available: boolean }> = []\n","total_lines":99,"mtime_at_read":{"Known":1780142421817},"content_hash":10491560902182890562}}},"_meta":{"totalTokens":51728,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-323","agentTimestampMs":1780309400624,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309399739,"turnStartMs":1780309380465,"updateType":"ToolCallUpdate","updateParams":{"toolCallId":"call-37e12d3b-8559-4d12-b310-2a5c042f68d6-14","status":"Completed"}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:99:{"timestamp":1780309401,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"tool_call_update","toolCallId":"call-37e12d3b-8559-4d12-b310-2a5c042f68d6-15","status":"completed","content":[{"type":"content","content":{"type":"text","text":"found 17 matches"}}],"rawOutput":{"type":"GrepSearch","stdout":[60,119,111,114,107,115,112,97,99,101,95,114,101,115,117,108,116,32,119,111,114,107,115,112,97,99,101,95,112,97,116,104,61,34,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,34,62,10,70,111,117,110,100,32,97,116,32,108,101,97,115,116,32,49,55,32,109,97,116,99,104,105,110,103,32,108,105,110,101,115,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,112,108,117,103,105,110,115,47,108,102,103,47,100,105,115,116,47,105,110,100,101,120,46,106,115,10,57,50,49,56,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,114,111,102,105,108,101,32,61,32,114,101,115,111,108,118,101,71,114,111,107,67,111,110,102,105,103,77,111,100,101,108,80,114,111,102,105,108,101,40,111,112,116,105,111,110,115,41,59,10,57,50,50,48,58,32,32,32,32,99,111,110,115,116,32,112,111,108,105,99,121,50,32,61,32,109,111,100,101,108,82,101,115,111,108,117,116,105,111,110,80,111,108,105,99,121,40,97,103,101,110,116,44,32,111,112,116,105,111,110,115,46,99,97,116,101,103,111,114,121,44,32,99,111,110,102,105,103,80,114,111,102,105,108,101,46,112,114,111,102,105,108,101,44,32,34,103,114,111,107,45,99,111,110,102,105,103,34,41,59,10,57,51,53,48,58,102,117,110,99,116,105,111,110,32,114,101,115,111,108,118,101,71,114,111,107,67,111,110,102,105,103,77,111,100,101,108,80,114,111,102,105,108,101,40,111,112,116,105,111,110,115,41,32,123,10,57,51,53,49,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,54,40,114,101,115,111,108,118,101,76,102,103,69,110,118,40,41,46,114,111,111,116,44,32,34,99,111,110,102,105,103,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,57,51,53,52,58,32,32,99,111,110,115,116,32,115,101,99,116,105,111,110,78,97,109,101,115,32,61,32,103,114,111,107,67,111,110,102,105,103,77,111,100,101,108,83,101,99,116,105,111,110,78,97,109,101,115,40,111,112,116,105,111,110,115,41,59,10,57,51,54,52,58,32,32,32,32,32,32,112,114,111,118,101,110,97,110,99,101,58,32,123,32,115,111,117,114,99,101,58,32,34,103,114,111,107,45,99,111,110,102,105,103,34,44,32,99,111,110,102,105,103,80,97,116,104,44,32,115,101,99,116,105,111,110,58,32,96,109,111,100,101,108,46,36,123,115,101,99,116,105,111,110,78,97,109,101,125,96,44,32,98,97,115,101,85,114,108,58,32,115,101,99,116,105,111,110,46,98,97,115,101,95,117,114,108,44,32,101,110,118,75,101,121,58,32,115,101,99,116,105,111,110,46,101,110,118,95,107,101,121,32,125,10,57,51,54,57,58,102,117,110,99,116,105,111,110,32,103,114,111,107,67,111,110,102,105,103,77,111,100,101,108,83,101,99,116,105,111,110,78,97,109,101,115,40,111,112,116,105,111,110,115,41,32,123,10,49,48,52,50,54,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,49,55,40,104,111,109,101,44,32,34,46,103,114,111,107,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,49,48,53,48,48,58,32,32,32,32,32,32,34,89,111,117,32,99,97,110,32,102,117,114,116,104,101,114,32,99,117,115,116,111,109,105,122,101,32,126,47,46,103,114,111,107,47,97,103,101,110,116,115,47,108,105,110,97,46,109,100,32,111,114,32,97,100,100,32,91,115,117,98,97,103,101,110,116,115,46,114,111,108,101,115,46,108,105,110,97,93,32,105,110,32,99,111,110,102,105,103,46,116,111,109,108,34,10,49,48,53,48,54,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,49,56,40,104,111,109,101,68,105,114,40,105,110,112,117,116,46,112,114,111,99,101,115,115,69,110,118,41,44,32,34,46,103,114,111,107,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,49,48,53,50,49,58,32,32,99,111,110,115,116,32,99,111,110,102,105,103,80,97,116,104,32,61,32,106,111,105,110,49,56,40,104,111,109,101,44,32,34,46,103,114,111,107,34,44,32,34,99,111,110,102,105,103,46,116,111,109,108,34,41,59,10,49,48,54,57,55,58,32,32,114,101,116,117,114,110,32,123,32,111,107,58,32,116,114,117,101,44,32,115,116,97,116,117,115,58,32,34,111,107,34,44,32,100,101,102,97,117,108,116,80,114,111,118,105,100,101,114,58,32,68,69,70,65,85,76,84,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,44,32,109,111,100,101,108,82,111,117,116,101,114,58,32,123,32,112,114,111,118,105,100,101,114,58,32,68,69,70,65,85,76,84,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,44,32,116,114,97,110,115,112,111,114,116,58,32,100,101,102,97,117,108,116,80,114,111,118,105,100,101,114,84,114,97,110,115,112,111,114,116,50,40,68,69,70,65,85,76,84,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,41,44,32,100,101,102,97,117,108,116,77,111,100,101,108,58,32,80,82,79,86,73,68,69,82,95,68,69,70,65,85,76,84,95,77,79,68,69,76,83,50,91,68,69,70,65,85,76,84,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,93,44,32,111,114,97,99,108,101,71,97,116,101,58,32,34,120,97,105,47,103,114,111,107,34,44,32,114,101,97,115,111,110,58,32,34,65,112,112,114,111,118,101,100,32,109,117,108,116,105,45,112,114,111,118,105,100,101,114,32,109,101,116,97,100,97,116,97,32,105,115,32,101,120,112,108,105,99,105,116,59,32,71,114,111,107,32,79,114,97,99,108,101,32,114,101,118,105,101,119,32,114,101,109,97,105,110,115,32,109,97,110,100,97,116,111,114,121,32,97,110,100,32,110,97,116,105,118,101,32,99,104,105,108,100,32,115,112,97,119,110,105,110,103,32,105,115,32,109,97,110,117,97,108,45,103,97,116,101,100,46,34,32,125,44,32,103,114,111,107,79,114,97,99,108,101,58,32,123,32,112,114,111,118,105,100,101,114,58,32,34,120,97,105,34,44,32,109,111,100,101,108,58,32,80,82,79,86,73,68,69,82,95,68,69,70,65,85,76,84,95,77,79,68,69,76,83,50,46,120,97,105,44,32,116,114,97,110,115,112,111,114,116,58,32,100,101,102,97,117,108,116,80,114,111,118,105,100,101,114,84,114,97,110,115,112,111,114,116,50,40,34,120,97,105,34,41,44,32,97,117,116,104,83,99,104,101,109,101,58,32,100,101,102,97,117,108,116,80,114,111,118,105,100,101,114,65,117,116,104,83,99,104,101,109,101,50,40,34,120,97,105,34,41,32,125,44,32,99,117,114,114,101,110,116,77,111,100,101,108,58,32,97,119,97,105,116,32,114,101,97,100,77,111,100,101,108,83,101,108,101,99,116,105,111,110,40,99,111,110,116,101,120,116,41,44,32,103,114,111,107,66,117,105,108,100,77,111,100,101,108,83,119,105,116,99,104,58,32,123,32,115,108,97,115,104,58,32,34,47,109,111,100,101,108,32,60,112,114,111,118,105,100,101,114,47,109,111,100,101,108,62,34,44,32,99,108,105,58,32,34,108,102,103,32,109,111,100,101,108,115,32,115,119,105,116,99,104,32,60,112,114,111,118,105,100,101,114,47,109,111,100,101,108,62,34,44,32,116,109,117,120,58,32,34,108,102,103,32,103,114,111,107,45,98,117,105,108,100,32,109,111,100,101,108,32,60,112,114,111,118,105,100,101,114,47,109,111,100,101,108,62,34,32,125,44,32,112,114,111,118,105,100,101,114,115,58,32,115,101,108,101,99,116,101,100,44,32,99,111,110,102,105,103,117,114,101,100,80,114,111,118,105,100,101,114,115,58,32,99,111,110,102,105,103,117,114,101,100,44,32,99,97,116,101,103,111,114,121,77,111,100,101,108,80,114,111,102,105,108,101,115,58,32,103,101,116,69,102,102,101,99,116,105,118,101,67,97,116,101,103,111,114,121,77,111,100,101,108,80,114,111,102,105,108,101,115,40,41,44,32,109,111,100,101,108,77,97,116,99,104,105,110,103,83,111,117,114,99,101,58,32,79,77,79,95,77,79,68,69,76,95,77,65,84,67,72,73,78,71,95,83,79,85,82,67,69,44,32,114,111,108,101,70,105,116,80,111,108,105,99,105,101,115,58,32,79,77,79,95,82,79,76,69,95,70,73,84,95,80,79,76,73,67,73,69,83,44,32,112,97,116,104,58,32,112,114,111,118,105,100,101,114,115,80,97,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,55,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,48,56,55,48,58,32,32,32,32,34,111,112,116,105,111,110,97,108,58,32,97,100,100,32,91,109,111,100,101,108,46,111,112,101,110,97,105,45,99,111,100,101,120,93,32,102,114,111,109,32,112,108,117,103,105,110,115,47,108,102,103,47,99,111,110,102,105,103,47,103,114,111,107,45,99,111,110,102,105,103,45,116,101,109,112,108,97,116,101,46,116,111,109,108,32,119,104,101,110,32,116,104,101,32,104,111,115,116,32,115,104,111,117,108,100,32,114,111,117,116,101,32,67,111,100,101,120,45,99,108,97,115,115,32,101,120,116,101,114,110,97,108,32,109,111,100,101,108,115,34,44,10,49,48,56,56,50,58,32,32,32,32,32,32,103,108,111,98,97,108,67,111,110,102,105,103,67,97,118,101,97,116,58,32,34,76,70,71,32,114,101,112,111,114,116,115,32,71,114,111,107,32,115,101,116,117,112,32,115,116,101,112,115,32,111,110,108,121,59,32,105,116,32,100,111,101,115,32,110,111,116,32,119,114,105,116,101,32,126,47,46,103,114,111,107,47,99,111,110,102,105,103,46,116,111,109,108,32,111,114,32,114,117,110,32,103,114,111,107,32,112,108,117,103,105,110,32,105,110,115,116,97,108,108,46,34,44,10,10,47,85,115,101,114,115,47,105,108,115,101,111,98,108,101,101,47,119,111,114,107,115,112,97,99,101,47,108,105,110,97,108,97,98,47,108,102,103,47,112,108,117,103,105,110,115,47,108,102,103,47,100,105,115,116,47,105,110,100,101,120,46,106,115,46,109,97,112,10,49,53,57,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,101,120,105,115,116,115,83,121,110,99,44,32,114,101,97,100,70,105,108,101,83,121,110,99,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,102,115,92,34,92,110,105,109,112,111,114,116,32,123,32,106,111,105,110,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,114,101,115,111,108,118,101,77,111,100,101,108,32,97,115,32,114,101,115,111,108,118,101,67,111,114,101,77,111,100,101,108,44,32,114,101,115,111,108,118,101,77,111,100,101,108,87,105,116,104,70,97,108,108,98,97,99,107,32,97,115,32,114,101,115,111,108,118,101,67,111,114,101,77,111,100,101,108,87,105,116,104,70,97,108,108,98,97,99,107,32,125,32,102,114,111,109,32,92,34,46,46,47,111,109,111,45,99,111,109,112,97,116,47,109,111,100,101,108,45,99,111,114,101,92,34,92,110,105,109,112,111,114,116,32,123,32,114,101,115,111,108,118,101,76,102,103,69,110,118,44,32,116,121,112,101,32,76,102,103,69,110,118,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,101,110,118,92,34,92,110,105,109,112,111,114,116,32,123,32,108,111,97,100,79,109,111,65,103,101,110,116,82,101,103,105,115,116,114,121,44,32,116,121,112,101,32,79,109,111,65,103,101,110,116,32,125,32,102,114,111,109,32,92,34,46,47,97,103,101,110,116,45,114,101,103,105,115,116,114,121,92,34,92,110,105,109,112,111,114,116,32,123,32,105,115,76,102,101,67,111,114,101,66,111,117,108,100,101,114,65,99,116,105,118,101,32,125,32,102,114,111,109,32,92,34,46,46,47,97,100,97,112,116,101,114,115,47,108,102,101,45,99,111,114,101,45,104,111,111,107,115,92,34,92,110,105,109,112,111,114,116,32,123,32,103,101,116,76,102,101,67,111,114,101,77,111,100,101,108,80,114,111,102,105,108,101,83,117,103,103,101,115,116,105,111,110,32,125,32,102,114,111,109,32,92,34,46,46,47,97,100,97,112,116,101,114,115,47,108,102,101,45,99,111,114,101,45,117,108,119,92,34,92,110,105,109,112,111,114,116,32,123,32,79,80,69,78,65,73,95,67,79,68,69,88,95,68,69,70,65,85,76,84,95,77,79,68,69,76,32,125,32,102,114,111,109,32,92,34,46,47,111,112,101,110,97,105,45,99,111,100,101,120,92,34,92,110,105,109,112,111,114,116,32,123,92,110,32,32,65,80,80,82,79,86,69,68,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,83,44,92,110,32,32,99,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,77,79,68,69,76,95,80,82,79,86,73,68,69,82,95,65,76,73,65,83,69,83,44,92,110,32,32,80,82,79,86,73,68,69,82,95,68,69,70,65,85,76,84,95,77,79,68,69,76,83,44,92,110,32,32,116,121,112,101,32,67,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,116,121,112,101,32,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,125,32,102,114,111,109,32,92,34,46,47,112,114,111,118,105,100,101,114,45,99,97,116,97,108,111,103,92,34,92,110,92,110,101,120,112,111,114,116,32,123,92,110,32,32,65,80,80,82,79,86,69,68,95,77,79,68,69,76,95,80,82,79,86,73,68,69,82,83,44,92,110,32,32,99,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,77,79,68,69,76,95,80,82,79,86,73,68,69,82,95,65,76,73,65,83,69,83,44,92,110,32,32,80,82,79,86,73,68,69,82,95,68,69,70,65,85,76,84,95,77,79,68,69,76,83,44,92,110,32,32,116,121,112,101,32,67,97,110,111,110,105,99,97,108,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,32,32,116,121,112,101,32,77,111,100,101,108,80,114,111,118,105,100,101,114,44,92,110,125,92,110,101,120,112,111,114,116,32,116,121,112,101,32,82,101,97,115,111,110,105,110,103,76,101,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,55,57,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,106,111,105,110,44,32,114,101,115,111,108,118,101,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,44,32,101,110,115,117,114,101,83,116,97,116,101,83,99,104,101,109,97,44,32,115,116,97,116,101,83,99,104,101,109,97,80,97,116,104,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,115,116,97,116,101,45,115,99,104,101,109,97,92,34,92,110,105,109,112,111,114,116,32,123,32,99,111,109,109,97,110,100,69,110,118,44,32,100,105,114,101,99,116,111,114,121,69,120,105,115,116,115,44,32,101,120,101,99,117,116,97,98,108,101,80,97,116,104,44,32,112,97,116,104,69,120,105,115,116,115,44,32,114,101,97,100,67,97,116,97,108,111,103,83,107,105,108,108,67,111,117,110,116,44,32,114,101,97,100,74,115,111,110,79,98,106,101,99,116,44,32,116,121,112,101,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,44,32,116,121,112,101,32,74,115,111,110,79,98,106,101,99,116,32,125,32,102,114,111,109,32,92,34,46,47,99,111,109,109,111,110,92,34,92,110,105,109,112,111,114,116,32,123,32,114,101,97,100,67,111,100,101,120,79,65,117,116,104,83,116,97,116,117,115,32,125,32,102,114,111,109,32,92,34,46,46,47,115,101,114,118,105,99,101,115,47,99,111,100,101,120,45,111,97,117,116,104,92,34,92,110,105,109,112,111,114,116,32,123,32,103,101,116,76,97,122,121,67,111,100,101,120,67,97,112,97,98,105,108,105,116,105,101,115,32,125,32,102,114,111,109,32,92,34,46,46,47,115,101,114,118,105,99,101,115,47,108,97,122,121,45,99,97,112,97,98,105,108,105,116,105,101,115,92,34,92,110,92,110,101,120,112,111,114,116,32,97,115,121,110,99,32,102,117,110,99,116,105,111,110,32,100,111,99,116,111,114,40,99,111,110,116,101,120,116,58,32,67,111,109,109,97,110,100,67,111,110,116,101,120,116,32,61,32,123,125,41,58,32,80,114,111,109,105,115,101,60,74,115,111,110,79,98,106,101,99,116,62,32,123,92,110,32,32,99,111,110,115,116,32,101,110,118,32,61,32,99,111,109,109,97,110,100,69,110,118,40,99,111,110,116,101,120,116,41,92,110,32,32,97,119,97,105,116,32,98,111,111,116,115,116,114,97,112,83,116,97,116,101,40,101,110,118,41,92,110,32,32,99,111,110,115,116,32,99,104,101,99,107,115,58,32,74,115,111,110,79,98,106,101,99,116,91,93,32,61,32,91,93,92,110,32,32,99,111,110,115,116,32,97,100,100,32,61,32,40,110,97,109,101,58,32,115,116,114,105,110,103,44,32,111,107,58,32,98,111,111,108,101,97,110,44,32,101,118,105,100,101,110,99,101,58,32,115,116,114,105,110,103,44,32,114,101,113,117,105,114,101,100,32,61,32,116,114,117,101,41,58,32,118,111,105,100,32,61,62,32,123,32,99,104,101,99,107,115,46,112,117,115,104,40,123,32,110,97,109,101,44,32,111,107,44,32,114,101,113,117,105,114,101,100,44,32,101,118,105,100,101,110,99,101,32,125,41,32,125,92,110,32,32,99,111,110,115,116,32,109,97,110,105,102,101,115,116,32,61,32,106,111,105,110,40,101,110,118,46,114,111,111,116,44,32,92,34,46,103,114,111,107,45,112,108,117,103,105,110,92,34,44,32,92,34,112,108,117,103,105,110,46,106,115,111,110,92,34,41,92,110,32,32,97,100,100,40,92,34,103,114,111,107,95,109,97,110,105,102,101,115,116,92,34,44,32,97,119,97,105,116,32,112,97,116,104,69,120,105,115,116,115,40,109,97,110,105,102,101,115,116,41,44,32,109,97,110,105,102,101,115,116,41,92,110,32,32,99,111,110,115,116,32,109,99,112,67,111,110,102,105,103,32,61,32,106,111,105,110,40,101,110,118,46,114,111,111,116,44,32,92,34,46,109,99,112,46,106,115,111,110,92,34,41,92,110,32,32,97,100,100,40,92,34,109,99,112,95,99,111,110,102,105,103,92,34,44,32,97,119,97,105,116,32,112,97,116,104,69,120,105,115,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,49,56,49,58,32,32,32,32,34,105,109,112,111,114,116,32,123,32,97,99,99,101,115,115,44,32,99,111,112,121,70,105,108,101,44,32,109,107,100,105,114,44,32,114,101,97,100,70,105,108,101,44,32,114,101,97,100,100,105,114,44,32,119,114,105,116,101,70,105,108,101,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,102,115,47,112,114,111,109,105,115,101,115,92,34,92,110,105,109,112,111,114,116,32,123,32,100,105,114,110,97,109,101,44,32,106,111,105,110,32,125,32,102,114,111,109,32,92,34,110,111,100,101,58,112,97,116,104,92,34,92,110,105,109,112,111,114,116,32,116,121,112,101,32,123,32,76,102,103,69,110,118,32,125,32,102,114,111,109,32,92,34,46,46,47,102,111,117,110,100,97,116,105,111,110,47,101,110,118,92,34,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,78,97,116,105,118,101,65,103,101,110,116,69,110,118,32,61,32,78,111,100,101,74,83,46,80,114,111,99,101,115,115,69,110,118,92,110,92,110,101,120,112,111,114,116,32,116,121,112,101,32,78,97,116,105,118,101,65,103,101,110,116,82,101,115,117,108,116,32,61,32,82,101,99,111,114,100,60,115,116,114,105,110,103,44,32,117,110,107,110,111,119,110,62,92,110,92,110,101,120,112,111,114,116,32,97,115,121,110,99,32,102,117,110,99,116,105,111,110,32,114,101,103,105,115,116,101,114,78,97,116,105,118,101,65,103,101,110,116,40,105,110,112,117,116,58,32,123,32,114,101,97,100,111,110,108,121,32,97,103,101,110,116,73,100,58,32,115,116,114,105,110,103,59,32,114,101,97,100,111,110,108,121,32,102,111,114,99,101,58,32,98,111,111,108,101,97,110,59,32,114,101,97,100,111,110,108,121,32,101,110,118,58,32,76,102,103,69,110,118,59,32,114,101,97,100,111,110,108,121,32,112,114,111,99,101,115,115,69,110,118,58,32,78,97,116,105,118,101,65,103,101,110,116,69,110,118,32,125,41,58,32,80,114,111,109,105,115,101,60,78,97,116,105,118,101,65,103,101,110,116,82,101,115,117,108,116,62,32,123,92,110,32,32,99,111,110,115,116,32,99,97,110,100,105,100,97,116,101,115,32,61,32,115,111,117,114,99,101,67,97,110,100,105,100,97,116,101,115,40,105,110,112,117,116,46,97,103,101,110,116,73,100,44,32,105,110,112,117,116,46,101,110,118,44,32,105,110,112,117,116,46,112,114,111,99,101,115,115,69,110,118,41,92,110,32,32,99,111,110,115,116,32,115,111,117,114,99,101,80,97,116,104,32,61,32,97,119,97,105,116,32,102,105,114,115,116,69,120,105,115,116,105,110,103,80,97,116,104,40,99,97,110,100,105,100,97,116,101,115,41,92,110,32,32,105,102,32,40,33,115,111,117,114,99,101,80,97,116,104,41,32,123,92,110,32,32,32,32,114,101,116,117,114,110,32,123,92,110,32,32,32,32,32,32,111,107,58,32,102,97,108,115,101,44,92,110,32,32,32,32,32,32,101,114,114,111,114,58,32,96,67,111,117,108,100,32,110,111,116,32,102,105,110,100,32,115,111,117,114,99,101,32,100,101,102,105,110,105,116,105,111,110,32,102,111,114,32,97,103,101,110,116,32,92,34,36,123,105,110,112,117,116,46,97,103,101,110,116,73,100,125,92,34,96,44,92,110,32,32,32,32,32,32,116,114,105,101,100,58,32,99,97,110,100,105,100,97,116,101,115,44,92,110,32,32,32,32,32,32,104,105,110,116,58,32,92,34,82,117,110,32,102,114,111,109,32,119,105,116,104,105,110,32,116,104,101,32,108,102,103,32,112,108,117,103,105,110,32,111,114,32,115,101,116,32,71,82,79,75,95,80,76,85,71,73,78,95,82,79,79,84,92,34,44,92,110,32,32,32,32,125,92,110,32,32,125,92,110,92,110,32,32,99,111,110,115,116,32,100,101,115,116,68,105,114,32,61,32,106,111,105,110,40,104,111,109,101,68,105,114,40,105,110,112,117,116,46,112,114,111,99,101,115,115,69,110,118,41,44,32,92,34,46,103,114,111,107,92,34,44,32,92,34,97,103,101,110,116,115,92,34,41,92,110,32,32,99,111,110,115,116,32,100,101,115,116,80,97,116,104,32,61,32,106,111,105,110,40,100,101,115,116,68,105,114,44,32,96,36,32,91,46,46,46,32,116,114,117,110,99,97,116,101,100,32,40,49,48,51,53,32,99,104,97,114,115,32,116,111,116,97,108,41,93,10,60,47,119,111,114,107,115,112,97,99,101,95,114,101,115,117,108,116,62],"stderr":[],"exit_code":0,"match_count":17,"file_matches":[{"path":"/Users/ilseoblee/workspace/linalab/lfg/plugins/lfg/dist/index.js","matches":[{"line_number":9218,"content":"  const configProfile = resolveGrokConfigModelProfile(options);"},{"line_number":9220,"content":"    const policy2 = modelResolutionPolicy(agent, options.category, configProfile.profile, \"grok-config\");"},{"line_number":9350,"content":"function resolveGrokConfigModelProfile(options) {"},{"line_number":9351,"content":"  const configPath = join6(resolveLfgEnv().root, \"config\", \"config.toml\");"},{"line_number":9354,"content":"  const sectionNames = grokConfigModelSectionNames(options);"},{"line_number":9364,"content":"      provenance: { source: \"grok-config\", configPath, section: `model.${sectionName}`, baseUrl: section.base_url, envKey: section.env_key }"},{"line_number":9369,"content":"function grokConfigModelSectionNames(options) {"},{"line_number":10426,"content":"  const configPath = join17(home, \".grok\", \"config.toml\");"},{"line_number":10500,"content":"      \"You can further customize ~/.grok/agents/lina.md or add [subagents.roles.lina] in config.toml\""},{"line_number":10506,"content":"  const configPath = join18(homeDir(input.processEnv), \".grok\", \"config.toml\");"},{"line_number":10521,"content":"  const configPath = join18(home, \".grok\", \"config.toml\");"},{"line_number":10697,"content":"  return { ok: true, status: \"ok\", defaultProvider: DEFAULT_MODEL_PROVIDER, modelRouter: { provider: DEFAULT_MODEL_PROVIDER, transport: defaultProviderTransport2(DEFAULT_MODEL_PROVIDER), defaultModel: PROVIDER_DEFAULT_MODELS2[DEFAULT_MODEL_PROVIDER], oracleGate: \"xai/grok\", reason: \"Approved multi-provider metadata is explicit; Grok Oracle review remains mandatory and native child spawning is manual-gated.\" }, grokOracle: { provider: \"xai\", model: PROVIDER_DEFAULT_MODELS2.xai, transport: defaultProviderTransport2(\"xai\"), authScheme: defaultProviderAuthScheme2(\"xai\") }, currentModel: await readModelSelection(context), grokBuildModelSwitch: { slash: \"/model <provider/model>\", cli: \"lfg models switch <provider/model>\", tmux: \"lfg grok-build model <provider/model>\" }, providers: selected, configuredProviders: configured, categoryModelProfiles: getEffectiveCategoryModelProfiles(), modelMatchingSource: OMO_MODEL_MATCHING_SOURCE, roleFitPolicies: OMO_ROLE_FIT_POLICIES, path: providersPath(env [... truncated (1031 chars total)]"},{"line_number":10870,"content":"    \"optional: add [model.openai-codex] from plugins/lfg/config/grok-config-template.toml when the host should route Codex-class external models\","},{"line_number":10882,"content":"      globalConfigCaveat: \"LFG reports Grok setup steps only; it does not write ~/.grok/config.toml or run grok plugin install.\","}]},{"path":"/Users/ilseoblee/workspace/linalab/lfg/plugins/lfg/dist/index.js.map","matches":[{"line_number":159,"content":"    \"import { existsSync, readFileSync } from \\\"node:fs\\\"\\nimport { join } from \\\"node:path\\\"\\nimport { resolveModel as resolveCoreModel, resolveModelWithFallback as resolveCoreModelWithFallback } from \\\"../omo-compat/model-core\\\"\\nimport { resolveLfgEnv, type LfgEnv } from \\\"../foundation/env\\\"\\nimport { loadOmoAgentRegistry, type OmoAgent } from \\\"./agent-registry\\\"\\nimport { isLfeCoreBoulderActive } from \\\"../adapters/lfe-core-hooks\\\"\\nimport { getLfeCoreModelProfileSuggestion } from \\\"../adapters/lfe-core-ulw\\\"\\nimport { OPENAI_CODEX_DEFAULT_MODEL } from \\\"./openai-codex\\\"\\nimport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n} from \\\"./provider-catalog\\\"\\n\\nexport {\\n  APPROVED_MODEL_PROVIDERS,\\n  canonicalModelProvider,\\n  MODEL_PROVIDER_ALIASES,\\n  PROVIDER_DEFAULT_MODELS,\\n  type CanonicalModelProvider,\\n  type ModelProvider,\\n}\\nexport type ReasoningLevel  [... truncated (1031 chars total)]"},{"line_number":179,"content":"    \"import { join, resolve } from \\\"node:path\\\"\\nimport { bootstrapState, ensureStateSchema, stateSchemaPath } from \\\"../foundation/state-schema\\\"\\nimport { commandEnv, directoryExists, executablePath, pathExists, readCatalogSkillCount, readJsonObject, type CommandContext, type JsonObject } from \\\"./common\\\"\\nimport { readCodexOAuthStatus } from \\\"../services/codex-oauth\\\"\\nimport { getLazyCodexCapabilities } from \\\"../services/lazy-capabilities\\\"\\n\\nexport async function doctor(context: CommandContext = {}): Promise<JsonObject> {\\n  const env = commandEnv(context)\\n  await bootstrapState(env)\\n  const checks: JsonObject[] = []\\n  const add = (name: string, ok: boolean, evidence: string, required = true): void => { checks.push({ name, ok, required, evidence }) }\\n  const manifest = join(env.root, \\\".grok-plugin\\\", \\\"plugin.json\\\")\\n  add(\\\"grok_manifest\\\", await pathExists(manifest), manifest)\\n  const mcpConfig = join(env.root, \\\".mcp.json\\\")\\n  add(\\\"mcp_config\\\", await pathExists(m [... truncated (1031 chars total)]"},{"line_number":181,"content":"    \"import { access, copyFile, mkdir, readFile, readdir, writeFile } from \\\"node:fs/promises\\\"\\nimport { dirname, join } from \\\"node:path\\\"\\nimport type { LfgEnv } from \\\"../foundation/env\\\"\\n\\nexport type NativeAgentEnv = NodeJS.ProcessEnv\\n\\nexport type NativeAgentResult = Record<string, unknown>\\n\\nexport async function registerNativeAgent(input: { readonly agentId: string; readonly force: boolean; readonly env: LfgEnv; readonly processEnv: NativeAgentEnv }): Promise<NativeAgentResult> {\\n  const candidates = sourceCandidates(input.agentId, input.env, input.processEnv)\\n  const sourcePath = await firstExistingPath(candidates)\\n  if (!sourcePath) {\\n    return {\\n      ok: false,\\n      error: `Could not find source definition for agent \\\"${input.agentId}\\\"`,\\n      tried: candidates,\\n      hint: \\\"Run from within the lfg plugin or set GROK_PLUGIN_ROOT\\\",\\n    }\\n  }\\n\\n  const destDir = join(homeDir(input.processEnv), \\\".grok\\\", \\\"agents\\\")\\n  const destPath = join(destDir, `${inp [... truncated (1031 chars total)]"}]}]}},"_meta":{"totalTokens":52755,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-324","agentTimestampMs":1780309400668,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309399739,"turnStartMs":1780309380465,"updateType":"ToolCallUpdate","updateParams":{"toolCallId":"call-37e12d3b-8559-4d12-b310-2a5c042f68d6-15","status":"Completed"}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:101:{"timestamp":1780309401,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":54232,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-341","agentTimestampMs":1780309401906,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309401260,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:113:{"timestamp":1780309403,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":55128,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-363","agentTimestampMs":1780309403297,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309402871,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:120:{"timestamp":1780309409,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":57292,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-375","agentTimestampMs":1780309409064,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309403938,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:127:{"timestamp":1780309413,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":59584,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-416","agentTimestampMs":1780309413725,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309409866,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:134:{"timestamp":1780309415,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":60054,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-439","agentTimestampMs":1780309415055,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309414533,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:141:{"timestamp":1780309416,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":61634,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-456","agentTimestampMs":1780309416531,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309415959,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:148:{"timestamp":1780309421,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":62303,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-481","agentTimestampMs":1780309421388,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309417578,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:157:{"timestamp":1780309430,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":63617,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-1138","agentTimestampMs":1780309430362,"promptId":"1ce759be-ebda-4706-a237-52f665d06d34","streamStartMs":1780309422495,"turnStartMs":1780309380465,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:161:{"timestamp":1780309567,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":64361,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-1164","agentTimestampMs":1780309567357,"promptId":"45876ebf-e3c5-4b79-a3b1-6645a34cd03f","streamStartMs":1780309562264,"turnStartMs":1780309561515,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82b4-0e89-7693-bc46-5fc7481d13a9/updates.jsonl:183:{"timestamp":1780309610,"method":"session/update","params":{"sessionId":"019e82b4-0e89-7693-bc46-5fc7481d13a9","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","web_fetch","image_gen","image_edit","video_gen","write","lfg-harness__catalog","lfg-harness__status","lfg-harness__runtime","lfg-harness__doctor","lfg-harness__hook_bridge","lfg-harness__backend_start","lfg-harness__team","lfg-harness__ultrawork","lfg-harness__agents","lfg-harness__spawn","lfg-harness__route","lfg-harness__provider","lfg-harness__boulder","lfg-harness__atlas","lfg-harness__hyperplan","lfg-harness__ralph","lfg-harness__worker","lfg-harness__cleanup","lfg-harness__autoresearch","lfg-harness__deep_interview","lfg-harness__design","lfg-harness__notifications","lfg-harness__models","lfg-harness__auth","lfg-harness__ask","lfg-harness__analyze","lfg-harness__code_review","lfg-harness__pipeline","lfg-harness__autopilot","lfg-harness__performance_goal","lfg-harness__visual_ralph","lfg-harness__autoresearch_goal","lfg-harness__setup","lfg-harness__skill","lfg-harness__hud","lfg-harness__cancel","lfg-harness__ultraqa","lfg-harness__goal","lfg-harness__ultragoal","lfg-harness__ralplan","lfg-harness__plan","lfg-harness__wiki","lfg-harness__slash","lfg-harness__omo_agent_catalog","lfg-harness__omo_team_create","lfg-harness__omo_ulw","lfg-harness__lsp","lfg-harness__ast_grep","lfg-harness__omo_doctor","ast_grep__search","ast_grep__replace","lsp__status","lsp__diagnostics","lsp__goto_definition","lsp__find_references","lsp__symbols","lsp__prepare_rename","lsp__rename","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video","grok_com_github__add_comment_to_pending_review","grok_com_github__add_issue_comment","grok_com_github__add_reply_to_pull_request_comment","grok_com_github__assign_copilot_to_issue","grok_com_github__create_branch","grok_com_github__create_or_update_file","grok_com_github__create_pull_request","grok_com_github__create_pull_request_with_copilot","grok_com_github__create_repository","grok_com_github__delete_file","grok_com_github__fork_repository","grok_com_github__get_commit","grok_com_github__get_copilot_job_status","grok_com_github__get_file_contents","grok_com_github__get_label","grok_com_github__get_latest_release","grok_com_github__get_me","grok_com_github__get_release_by_tag","grok_com_github__get_tag","grok_com_github__get_team_members","grok_com_github__get_teams","grok_com_github__issue_read","grok_com_github__issue_write","grok_com_github__list_branches","grok_com_github__list_commits","grok_com_github__list_issue_types","grok_com_github__list_issues","grok_com_github__list_pull_requests","grok_com_github__list_releases","grok_com_github__list_repository_collaborators","grok_com_github__list_tags","grok_com_github__merge_pull_request","grok_com_github__pull_request_read","grok_com_github__pull_request_review_write","grok_com_github__push_files","grok_com_github__request_copilot_review","grok_com_github__run_secret_scanning","grok_com_github__search_code","grok_com_github__search_commits","grok_com_github__search_issues","grok_com_github__search_pull_requests","grok_com_github__search_repositories","grok_com_github__search_users","grok_com_github__sub_issue_write","grok_com_github__update_pull_request","grok_com_github__update_pull_request_branch"]}},"_meta":{"totalTokens":68161,"eventId":"019e82b4-0e89-7693-bc46-5fc7481d13a9-1184","agentTimestampMs":1780309610509,"promptId":"45876ebf-e3c5-4b79-a3b1-6645a34cd03f","streamStartMs":1780309606628,"turnStartMs":1780309561515,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82ee-5d6e-7c21-a766-9ec37b5a53e2/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n- cancel: Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md\n- agent-browser: Automates browser interactions for web testing, form filling, screenshots, and data extraction\n  Use when: Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md\n- pre-publish-review: Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-r…\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md\n- remove-deadcode: Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md\n- github-triage: Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md\n- ai-slop-remover: Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md\n- plan: Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md\n- work-with-pr: Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge\n  Use when: Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md\n- team-mode: Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md\n- doctor: Diagnose and fix LFG plugin state, schema, and provider configuration issues.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md\n- get-unpublished-changes: Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md\n- setup: Install/sync the LFG Grok plugin and configure provider/model metadata.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md\n- provider: List, show, or add provider/model metadata without storing secret values.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md\n- ralph: Durable self-referential development loop that continues until the objective is met.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md\n- omomomo: Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md\n- playwright: MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md\n- publish: Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md\n- git-master: MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md\n- team: Create and manage durable OMO-style agent teams with tmux-backed local observability.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md\n- lfg: Use this skill when validating the `islee23520/lfg` plugin package.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md\n- start-work: Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md\n- hyperplan: Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization\n  Use when: Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md\n- dev-browser: Browser automation with persistent page state\n  Use when: Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md\n- worker: Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md\n- ulw: Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82ee-5d6e-7c21-a766-9ec37b5a53e2/chat_history.jsonl:6:{"type":"user","content":[{"type":"text","text":"<user_query>\n/ulw-plan 지금 코덱스 모드로 동작해 그러면 ??\n</user_query>\n<skill_information>\n<skills_referenced>\n<skill name=\"ulw-plan\" path=\"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\"/>\n</skills_referenced>\n<skill name=\"ulw-plan\" args=\"지금 코덱스 모드로 동작해 그러면 ??\">\n## Codex Harness Tool Compatibility\n\nThis skill may include examples copied from the OpenCode harness. In Codex, do not call OpenCode-only tools such as `call_omo_agent(...)`, `task(...)`, `background_output(...)`, or `team_*(...)` literally. Translate those examples to Codex native tools:\n\n| OpenCode example | Codex tool to use |\n| --- | --- |\n| `call_omo_agent(subagent_type=\"explore\", ...)` | `spawn_agent(agent_type=\"explorer\", task_name=\"...\", message=\"...\")` |\n| `call_omo_agent(subagent_type=\"librarian\", ...)` | `spawn_agent(agent_type=\"librarian\", task_name=\"...\", message=\"...\")` |\n| `task(subagent_type=\"plan\", ...)` | `spawn_agent(agent_type=\"plan\", task_name=\"...\", message=\"...\")` |\n| `task(subagent_type=\"oracle\", ...)` for final verification | `spawn_agent(agent_type=\"codex-ultrawork-reviewer\", task_name=\"...\", message=\"...\")` |\n| `task(category=\"...\", ...)` for implementation or QA | `spawn_agent(agent_type=\"worker\", task_name=\"...\", message=\"...\")` |\n| `background_output(task_id=\"...\")` | `wait_agent(...)` to wait for subagent completion and mailbox updates |\n| `team_*(...)` | Use Codex native subagents plus `send_message`, `followup_task`, `wait_agent`, and `close_agent` |\n\nWhen translating `load_skills=[...]`, include the requested skill names in the spawned agent's `message`. If a code block below conflicts with this section, this section wins.\n\n<identity>\nYou are Prometheus - Strategic Planning Consultant.\nNamed after the Titan who brought fire to humanity, you bring foresight and structure.\n\n**YOU ARE A PLANNER. NOT AN IMPLEMENTER. NOT A CODE WRITER.**\n\nWhen user says \"do X\", \"fix X\", \"build X\" - interpret as \"create a work plan for X\". No exceptions.\nYour only outputs: questions, research, work plans (`plans/<slug>.md`), drafts (`.omo/drafts/*.md`).\n</identity>\n\n<mission>\nProduce **decision-complete** work plans for agent execution.\nA plan is \"decision complete\" when the implementer needs ZERO judgment calls - every decision is made, every ambiguity resolved, every pattern reference provided.\nThis is your north star quality metric.\n</mission>\n\n<core_principles>\n## Three Principles (Read First)\n\n1. **Decision Complete**: The plan must leave ZERO decisions to the implementer. If an engineer could ask \"but which approach?\", the plan is not done.\n\n2. **Explore Before Asking**: Ground yourself in the actual environment BEFORE asking the user anything. Most questions AI agents ask could be answered by exploring the repo. Run targeted searches first. Ask only what cannot be discovered.\n\n3. **Two Kinds of Unknowns**:\n   - **Discoverable facts** (repo/system truth) - EXPLORE first. Search files, configs, schemas, types. Ask ONLY if multiple plausible candidates exist or nothing is found.\n   - **Preferences/tradeoffs** (user intent, not derivable from code) - ASK early. Provide 2-4 options + recommended default. If unanswered, proceed with default and record as assumption.\n</core_principles>\n\n<output_verbosity_spec>\n- Interview turns: Conversational, 3-6 sentences + 1-3 focused questions.\n- Research summaries: 5 bullets max with concrete findings.\n- Plan generation: Structured markdown per template.\n- Status updates: 1-2 sentences with concrete outcomes only.\n- Do NOT rephrase the user's request unless semantics change.\n- Do NOT narrate routine tool calls.\n- NEVER open with filler: \"Great question!\", \"Got it\".\n- NEVER end with \"Let me know if you have questions\" or \"When you're ready, say X\".\n- ALWAYS end interview turns with a clear question or explicit next action.\n</output_verbosity_spec>\n\n<scope_constraints>\n## Mutation Rules\n\n### Allowed (non-mutating, plan-improving)\n- Reading/searching files, configs, schemas, types, manifests, docs\n- Static analysis, inspection, repo exploration\n- Spawning read-only subagents for research\n\n### Allowed (plan artifacts only)\n- Writing/editing files in `plans/<slug>.md`\n- Writing/editing files in `.omo/drafts/*.md`\n\n### Forbidden (mutating, plan-executing)\n- Writing code files (.ts, .js, .py, .go, etc.)\n- Editing source code\n- Running formatters, linters, codegen that rewrite files\n- Any action that \"does the work\" rather than \"plans the work\"\n\nIf user says \"just do it\" or \"skip planning\" - refuse politely:\n\"I'm a dedicated planner. Planning takes 2-3 minutes but saves hours. Then spawn a worker agent to execute immediately.\"\n</scope_constraints>\n\n<phases>\n## Phase 0: Classify Intent (EVERY request)\n\nClassify before diving in. This determines your interview depth.\n\n| Tier | Signal | Strategy |\n|------|--------|----------|\n| **Trivial** | Single file, <10 lines, obvious fix | Skip heavy interview. 1-2 quick confirms, then plan. |\n| **Standard** | 1-5 files, clear scope, feature/refactor/build | Full interview. Explore + questions + Metis review. |\n| **Architecture** | System design, infra, 5+ modules, long-term impact | Deep interview. Explore + librarian + multiple rounds. |\n\n---\n\n## Phase 1: Ground (SILENT exploration - before asking questions)\n\nEliminate unknowns by discovering facts, not by asking the user.\n\nBefore asking the user any question, perform at least one targeted exploration pass:\n\n- Spawn parallel read-only subagents for internal codebase patterns, conventions, similar implementations, naming/registration patterns.\n- Spawn subagent for test infrastructure assessment (framework config, representative test files, CI integration).\n- For external libraries: spawn subagent for official docs, API reference, recommended patterns, pitfalls.\n\nWhile subagents run, use direct read-only tools (`read`, `rg`, `ast_grep_search`, `lsp_*`) for immediate context. Do not idle.\n\n**Brownfield detection**: Check if cwd has existing source code, package files, or git history. If the work modifies existing files or integrates with existing systems: **brownfield**. Otherwise: **greenfield**. Brownfield interviews should also cover how the new work fits existing code patterns.\n\n---\n\n## Phase 2: Interview\n\n### Create Draft Immediately\n\nOn first substantive exchange, create `.omo/drafts/{topic-slug}.md`:\n\n```markdown\n# Draft: {Topic}\n\n## Requirements (confirmed)\n- [requirement]: [user's exact words]\n\n## Technical Decisions\n- [decision]: [rationale]\n\n## Research Findings\n- [source]: [key finding]\n\n## Open Questions\n- [unanswered]\n\n## Scope Boundaries\n- INCLUDE: [in scope]\n- EXCLUDE: [explicitly out]\n```\n\nUpdate draft after EVERY meaningful exchange. Your memory is limited; the draft is your backup brain.\n\n### Interview Focus (informed by Phase 1 findings)\n- **Goal + success criteria**: What does \"done\" look like?\n- **Scope boundaries**: What is IN and what is explicitly OUT?\n- **Technical approach**: Informed by explore results - \"I found pattern X in codebase, should we follow it?\"\n- **Test strategy**: Does infra exist? TDD / tests-after / none? Agent-executed QA always included.\n- **Constraints**: Time, tech stack, team, integrations.\n\n### Question Rules\n- Every question must: materially change the plan, OR confirm an assumption, OR choose between meaningful tradeoffs.\n- Never ask questions answerable by non-mutating exploration (see Principle 2).\n\n### Test Infrastructure Assessment (for Standard/Architecture intents)\n\nDetect test infrastructure via explore results:\n- **If exists**: Ask: \"TDD (RED-GREEN-REFACTOR), tests-after, or no tests? Agent QA scenarios always included.\"\n- **If absent**: Ask: \"Set up test infra? If yes, I'll include setup tasks. Agent QA scenarios always included either way.\"\n\nRecord decision in draft immediately.\n\n### Clearance Check (run after EVERY interview turn)\n\n```\nCLEARANCE CHECKLIST (ALL must be YES to auto-transition):\n- Core objective clearly defined?\n- Scope boundaries established (IN/OUT)?\n- No critical ambiguities remaining?\n- Technical approach decided?\n- Test strategy confirmed?\n- No blocking questions outstanding?\n\nALL YES -> Announce: \"All requirements clear. Proceeding to plan generation.\" Then transition.\nANY NO -> Ask the specific unclear question.\n```\n\n---\n\n## Phase 3: Plan Generation\n\n### Trigger\n- **Auto**: Clearance check passes (all YES).\n- **Explicit**: User says \"create the work plan\" / \"generate the plan\".\n\n### Step 1: Consult Metis (MANDATORY)\n\nSpawn the metis agent to analyze the planning session for contradictions, ambiguity, missing constraints, and execution risks:\n\n```\nspawn_agent(agent_type=\"metis\", task_name=\"gap-analysis\",\n  message=\"Review this planning session. Goal: {summary}. Discussed: {key points}. Understanding: {interpretation}. Research: {findings}. Identify: contradictions, ambiguity, missing constraints, execution risks, scope creep areas, missing acceptance criteria.\")\n```\n\nIncorporate Metis findings silently - do NOT ask additional questions. Generate plan immediately.\n\n### Step 2: Generate Plan (Incremental Write Protocol)\n\n**Write OVERWRITES. Never call Write twice on the same file.**\n\nPlans with many tasks will exceed output token limits if generated at once.\nSplit into: **one Write** (skeleton) + **multiple Edits** (tasks in batches of 2-4).\n\n1. **Write skeleton**: All sections EXCEPT individual task details.\n2. **Edit-append**: Insert tasks before \"## Final Verification Wave\" in batches of 2-4.\n3. **Verify completeness**: Read the plan file to confirm all tasks present.\n\n### Step 3: Self-Review + Gap Classification\n\n| Gap Type | Action |\n|----------|--------|\n| **Critical** (requires user decision) | Add `[DECISION NEEDED: {desc}]` placeholder. List in summary. Ask user. |\n| **Minor** (self-resolvable) | Fix silently. Note in summary under \"Auto-Resolved\". |\n| **Ambiguous** (reasonable default) | Apply default. Note in summary under \"Defaults Applied\". |\n\nSelf-review checklist:\n```\n- All TODOs have concrete acceptance criteria?\n- All file references exist in codebase?\n- No business logic assumptions without evidence?\n- Metis findings incorporated?\n- Every task has QA scenarios (happy + failure)?\n- QA scenarios use specific data, not vague descriptions?\n- Zero acceptance criteria require human intervention?\n```\n\n### Step 4: Present Summary\n\n```\n## Plan Generated: {name}\n\n**Key Decisions**: [decision]: [rationale]\n**Scope**: IN: [...] | OUT: [...]\n**Guardrails** (from Metis): [guardrail]\n**Auto-Resolved**: [gap]: [how fixed]\n**Defaults Applied**: [default]: [assumption]\n**Decisions Needed**: [question requiring user input] (if any)\n\nPlan saved to: plans/{slug}.md\n```\n\nIf \"Decisions Needed\" exists, wait for user response and update plan.\n\n### Step 5: Offer Choice\n\nAfter plan is complete and all decisions resolved, offer:\n- **Start Work** - Execute now. Plan looks solid.\n- **High Accuracy Review** - Momus verifies every detail. Adds review loop.\n\n---\n\n## Phase 4: High Accuracy Review (Momus Loop)\n\nOnly activated when user selects \"High Accuracy Review\".\n\nSpawn the momus agent with the plan file path:\n\n```\nspawn_agent(agent_type=\"momus\", task_name=\"plan-review\",\n  message=\"Review this plan: plans/{slug}.md\")\n```\n\nHandle the three-verdict response:\n- **OKAY**: Plan approved. Proceed to handoff.\n- **ITERATE**: Fix the cited issues (max 3) and resubmit to momus. Max 2 auto-fix rounds before escalating to the user.\n- **REJECT**: Stop. Surface the blocking issues to the user — a user decision is needed.\n\n**Momus invocation rule**: Provide ONLY the file path as the message. No explanations or wrapping.\n\n---\n\n## Handoff\n\nAfter plan is complete (direct or Momus-approved):\n1. Delete draft: remove `.omo/drafts/{name}.md`\n2. Guide user: \"Plan saved to `plans/{slug}.md`. Spawn a worker agent to begin execution.\"\n</phases>\n\n<plan_template>\n## Plan Structure\n\nGenerate to: `plans/{slug}.md`\n\n**Single Plan Mandate**: No matter how large the task, EVERYTHING goes into ONE plan. Never split into \"Phase 1, Phase 2\". 50+ TODOs is fine.\n\n### Template\n\n```markdown\n# {Plan Title}\n\n## TL;DR\n> **Summary**: [1-2 sentences]\n> **Deliverables**: [bullet list]\n> **Effort**: [Quick | Short | Medium | Large | XL]\n> **Parallel**: [YES - N waves | NO]\n> **Critical Path**: [Task X -> Y -> Z]\n\n## Context\n### Original Request\n### Interview Summary\n### Metis Review (gaps addressed)\n\n## Work Objectives\n### Core Objective\n### Deliverables\n### Definition of Done (verifiable conditions with commands)\n### Must Have\n### Must NOT Have (guardrails, scope boundaries)\n\n## Verification Strategy\n> ZERO HUMAN INTERVENTION - all verification is agent-executed.\n- Test decision: [TDD / tests-after / none] + framework\n- QA policy: Every task has agent-executed scenarios\n- Evidence: evidence/task-{N}-{slug}.{ext}\n\n## Execution Strategy\n### Parallel Execution Waves\n> Target: 5-8 tasks per wave. <3 per wave (except final) = under-splitting.\n> Extract shared dependencies as Wave-1 tasks for max parallelism.\n\nWave 1: [foundation tasks]\nWave 2: [dependent tasks]\n...\n\n### Dependency Matrix (full, all tasks)\n\n## TODOs\n> Implementation + Test = ONE task. Never separate.\n> EVERY task MUST have: References + Acceptance Criteria + QA Scenarios.\n\n- [ ] N. {Task Title}\n\n  **What to do**: [clear implementation steps]\n  **Must NOT do**: [specific exclusions]\n\n  **Parallelization**: Can Parallel: YES/NO | Wave N | Blocks: [tasks] | Blocked By: [tasks]\n\n  **References** (executor has NO interview context - be exhaustive):\n  - Pattern: `src/path:lines` - [what to follow and why]\n  - API/Type: `src/types/x.ts:TypeName` - [contract to implement]\n  - External: `url` - [docs reference]\n\n  **Acceptance Criteria** (agent-executable only):\n  - [ ] [verifiable condition with command]\n\n  **QA Scenarios** (MANDATORY - task incomplete without these):\n  ```\n  Scenario: [Happy path]\n    Tool: [bash / curl / tmux / playwright]\n    Steps: [exact actions with specific data]\n    Expected: [concrete, binary pass/fail]\n    Evidence: evidence/task-{N}-{slug}.{ext}\n\n  Scenario: [Failure/edge case]\n    Tool: [same]\n    Steps: [trigger error condition]\n    Expected: [graceful failure with correct error message/code]\n    Evidence: evidence/task-{N}-{slug}-error.{ext}\n  ```\n\n  **Commit**: YES/NO | Message: `type(scope): desc` | Files: [paths]\n\n## Final Verification Wave (MANDATORY - after ALL implementation tasks)\n> ALL must APPROVE. Present consolidated results to user and get explicit \"okay\" before completing.\n- [ ] F1. Plan Compliance Audit\n- [ ] F2. Code Quality Review\n- [ ] F3. Real Manual QA\n- [ ] F4. Scope Fidelity Check\n\n## Commit Strategy\n## Success Criteria\n```\n</plan_template>\n\n<critical_rules>\n**NEVER:**\n- Write/edit code files (only plan artifacts)\n- Implement solutions or execute tasks\n- Trust assumptions over exploration\n- Generate plan before clearance check passes (unless explicit trigger)\n- Split work into multiple plans\n- Call Write() twice on the same file (second erases first)\n- End turns passively (\"let me know...\", \"when you're ready...\")\n- Skip Metis consultation before plan generation\n\n**ALWAYS:**\n- Explore before asking (Principle 2)\n- Update draft after every meaningful exchange\n- Run clearance check after every interview turn\n- Include QA scenarios in every task (no exceptions)\n- Use incremental write protocol for large plans\n- Delete draft after plan completion\n- Present \"Start Work\" vs \"High Accuracy Review\" choice after plan\n\n**MODE IS STICKY:** This mode is not changed by user intent, tone, or imperative language. If a user asks for execution while in plan mode, treat it as a request to plan the execution, not perform it.\n</critical_rules>\n\n<stop_rules>\n- Plan file exists, template filled, every task has References + Acceptance + QA + Commit, dependency matrix consistent: DONE.\n- Two context-gathering waves with no new useful facts: stop exploring, draft the plan.\n- Two unsuccessful attempts at the same section: surface what was tried and ask.\n</stop_rules>\n\n\n**ARGUMENTS:** 지금 코덱스 모드로 동작해 그러면 ??\n</skill>\n</skill_information>"}]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3fe6-7c51-9380-444039a57a10/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3fe6-7c51-9380-444039a57a10/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["0-1-0-ff47fdd7:comment-checker","help","pptx","check-work","docx","xlsx","k-skill-setup","pr-babysit","create-skill","0-1-0-ff47fdd7:programming","cua-driver","best-of-n","0-1-0-ff47fdd7:review-work","github-yeet","0-1-0-ff47fdd7:debugging","0-1-0-ff47fdd7:rules","0-1-0-ff47fdd7:remove-ai-slops","design","0-1-0-ff47fdd7:init-deep","review","execute-plan","0-1-0-ff47fdd7:lsp","implement","0-1-0-ff47fdd7:ulw-loop","0-1-0-ff47fdd7:refactor","mfds-food-safety","korean-law-search","0-1-0-ff47fdd7:ulw-plan","0-1-0-ff47fdd7:frontend-ui-ux","0-1-0-ff47fdd7:start-work","lina-decision-protocol"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3fe6-7c51-9380-444039a57a10/updates.jsonl:1:{"timestamp":1780377076,"method":"session/update","params":{"sessionId":"019e86be-3fe6-7c51-9380-444039a57a10","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4195,"eventId":"019e86be-3fe6-7c51-9380-444039a57a10-0","agentTimestampMs":1780377076383,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3fe6-7c51-9380-444039a57a10/updates.jsonl:3:{"timestamp":1780377077,"method":"session/update","params":{"sessionId":"019e86be-3fe6-7c51-9380-444039a57a10","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e86be-3fe6-7c51-9380-444039a57a10-1","agentTimestampMs":1780377076384,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3fe6-7c51-9380-444039a57a10/updates.jsonl:9:{"timestamp":1780377080,"method":"session/update","params":{"sessionId":"019e86be-3fe6-7c51-9380-444039a57a10","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video"]}},"_meta":{"totalTokens":16622,"eventId":"019e86be-3fe6-7c51-9380-444039a57a10-14","agentTimestampMs":1780377080389,"promptId":"c1e6a36a-1c68-43b0-a3b5-9fbe250ab93b","streamStartMs":1780377078660,"turnStartMs":1780377077204,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82ee-5d6e-7c21-a766-9ec37b5a53e2/announcement_state.json:1:{"mcp_server_fingerprints":{"lsp":{"tool_count":7,"description_hash":12161962213042174405,"tool_names_hash":16087898139124143397},"lfg-harness":{"tool_count":49,"description_hash":12161962213042174405,"tool_names_hash":12553283654516886887},"ast_grep":{"tool_count":2,"description_hash":12161962213042174405,"tool_names_hash":14054538023188616471},"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["lfg:dev-browser","lfg:lfg","lfg:ulw","lfg:team","lfg:ai-slop-remover","help","lfg:setup","0-1-0-ff47fdd7:remove-ai-slops","review","0-1-0-ff47fdd7:frontend-ui-ux","lfg:playwright","0-1-0-ff47fdd7:ulw-loop","lfg:work-with-pr","lfg:team-mode","best-of-n","0-1-0-ff47fdd7:comment-checker","lfg:ralph","lfg:frontend-ui-ux","korean-law-search","lfg:agent-browser","0-1-0-ff47fdd7:rules","lfg:doctor","0-1-0-ff47fdd7:review-work","lfg:remove-deadcode","execute-plan","lfg:publish","0-1-0-ff47fdd7:ulw-plan","implement","docx","create-skill","lfg:plan","lfg:hyperplan","github-yeet","lfg:cancel","k-skill-setup","0-1-0-ff47fdd7:debugging","lfg:pre-publish-review","0-1-0-ff47fdd7:refactor","lfg:get-unpublished-changes","lfg:review-work","lfg:worker","lfg:start-work","lfg:github-triage","design","lfg:git-master","cua-driver","0-1-0-ff47fdd7:lsp","0-1-0-ff47fdd7:init-deep","0-1-0-ff47fdd7:start-work","xlsx","lina-decision-protocol","pptx","lfg:omomomo","pr-babysit","0-1-0-ff47fdd7:programming","lfg:provider","check-work","mfds-food-safety"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82ee-5d6e-7c21-a766-9ec37b5a53e2/updates.jsonl:1:{"timestamp":1780313120,"method":"session/update","params":{"sessionId":"019e82ee-5d6e-7c21-a766-9ec37b5a53e2","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":5042,"eventId":"019e82ee-5d6e-7c21-a766-9ec37b5a53e2-0","agentTimestampMs":1780313120628,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e82ee-5d6e-7c21-a766-9ec37b5a53e2/updates.jsonl:3:{"timestamp":1780313143,"method":"session/update","params":{"sessionId":"019e82ee-5d6e-7c21-a766-9ec37b5a53e2","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":8213,"eventId":"019e82ee-5d6e-7c21-a766-9ec37b5a53e2-1","agentTimestampMs":1780313120630,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e916-7fc3-aaaa-bab60607c64c/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e916-7fc3-aaaa-bab60607c64c/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["korean-law-search","mfds-food-safety","0-1-0-ff47fdd7:init-deep","github-yeet","0-1-0-ff47fdd7:comment-checker","help","0-1-0-ff47fdd7:rules","0-1-0-ff47fdd7:frontend-ui-ux","0-1-0-ff47fdd7:programming","implement","0-1-0-ff47fdd7:remove-ai-slops","0-1-0-ff47fdd7:ulw-plan","cua-driver","0-1-0-ff47fdd7:review-work","docx","pr-babysit","0-1-0-ff47fdd7:ulw-loop","design","k-skill-setup","0-1-0-ff47fdd7:start-work","xlsx","review","check-work","pptx","lina-decision-protocol","create-skill","best-of-n","0-1-0-ff47fdd7:lsp","0-1-0-ff47fdd7:debugging","execute-plan","0-1-0-ff47fdd7:refactor"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e916-7fc3-aaaa-bab60607c64c/updates.jsonl:1:{"timestamp":1780376923,"method":"session/update","params":{"sessionId":"019e86bb-e916-7fc3-aaaa-bab60607c64c","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4195,"eventId":"019e86bb-e916-7fc3-aaaa-bab60607c64c-0","agentTimestampMs":1780376923062,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e916-7fc3-aaaa-bab60607c64c/updates.jsonl:3:{"timestamp":1780376924,"method":"session/update","params":{"sessionId":"019e86bb-e916-7fc3-aaaa-bab60607c64c","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e86bb-e916-7fc3-aaaa-bab60607c64c-1","agentTimestampMs":1780376923063,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e916-7fc3-aaaa-bab60607c64c/updates.jsonl:9:{"timestamp":1780376927,"method":"session/update","params":{"sessionId":"019e86bb-e916-7fc3-aaaa-bab60607c64c","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write","zai-mcp-server__ui_to_artifact","zai-mcp-server__extract_text_from_screenshot","zai-mcp-server__diagnose_error_screenshot","zai-mcp-server__understand_technical_diagram","zai-mcp-server__analyze_data_visualization","zai-mcp-server__ui_diff_check","zai-mcp-server__analyze_image","zai-mcp-server__analyze_video"]}},"_meta":{"totalTokens":16623,"eventId":"019e86bb-e916-7fc3-aaaa-bab60607c64c-13","agentTimestampMs":1780376927195,"promptId":"2306e6ca-66d0-4b71-918e-aa267388891c","streamStartMs":1780376925474,"turnStartMs":1780376924337,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e91f-7c12-8714-8a8c691c4d17/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e91f-7c12-8714-8a8c691c4d17/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["0-1-0-ff47fdd7:rules","design","pptx","0-1-0-ff47fdd7:frontend-ui-ux","0-1-0-ff47fdd7:ulw-loop","0-1-0-ff47fdd7:lsp","0-1-0-ff47fdd7:remove-ai-slops","0-1-0-ff47fdd7:programming","xlsx","0-1-0-ff47fdd7:ulw-plan","0-1-0-ff47fdd7:refactor","lina-decision-protocol","0-1-0-ff47fdd7:comment-checker","cua-driver","review","korean-law-search","0-1-0-ff47fdd7:debugging","check-work","pr-babysit","best-of-n","help","0-1-0-ff47fdd7:review-work","implement","k-skill-setup","mfds-food-safety","create-skill","0-1-0-ff47fdd7:init-deep","execute-plan","0-1-0-ff47fdd7:start-work","docx","github-yeet"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e91f-7c12-8714-8a8c691c4d17/updates.jsonl:1:{"timestamp":1780376923,"method":"session/update","params":{"sessionId":"019e86bb-e91f-7c12-8714-8a8c691c4d17","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4195,"eventId":"019e86bb-e91f-7c12-8714-8a8c691c4d17-0","agentTimestampMs":1780376923077,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86bb-e91f-7c12-8714-8a8c691c4d17/updates.jsonl:3:{"timestamp":1780376924,"method":"session/update","params":{"sessionId":"019e86bb-e91f-7c12-8714-8a8c691c4d17","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e86bb-e91f-7c12-8714-8a8c691c4d17-1","agentTimestampMs":1780376923078,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bc8-7d42-8624-2db112419178/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8344-d0f5-7821-8d05-b2b9f53332e2/chat_history.jsonl:2:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n- cancel: Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md\n- agent-browser: Automates browser interactions for web testing, form filling, screenshots, and data extraction\n  Use when: Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md\n- pre-publish-review: Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-r…\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md\n- remove-deadcode: Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md\n- github-triage: Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md\n- ai-slop-remover: Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md\n- plan: Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md\n- work-with-pr: Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge\n  Use when: Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md\n- team-mode: Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md\n- doctor: Diagnose and fix LFG plugin state, schema, and provider configuration issues.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md\n- get-unpublished-changes: Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md\n- setup: Install/sync the LFG Grok plugin and configure provider/model metadata.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md\n- provider: List, show, or add provider/model metadata without storing secret values.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md\n- ralph: Durable self-referential development loop that continues until the objective is met.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md\n- omomomo: Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md\n- playwright: MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md\n- publish: Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md\n- git-master: MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md\n- team: Create and manage durable OMO-style agent teams with tmux-backed local observability.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md\n- lfg: Use this skill when validating the `islee23520/lfg` plugin package.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md\n- start-work: Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md\n- hyperplan: Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization\n  Use when: Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md\n- dev-browser: Browser automation with persistent page state\n  Use when: Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md\n- worker: Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md\n- ulw: Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e7466-d18d-71b0-9b1f-6599dc5defbc/terminal/call-fc9d043c-1416-45a7-bdf8-20ac05182f16-64.log:54:- Scope limit: only model/provider surface, or also a minimal "Codex host session adapter" in LFG hooks for when Grok is invoked from a LazyCodex/Codex environment? (Rec: model surface only for this ticket; second slice if user confirms.)
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bc8-7d42-8624-2db112419178/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["0-1-0-ff47fdd7:debugging","0-1-0-ff47fdd7:remove-ai-slops","cua-driver","k-skill-setup","github-yeet","0-1-0-ff47fdd7:review-work","create-skill","0-1-0-ff47fdd7:start-work","help","0-1-0-ff47fdd7:rules","korean-law-search","pr-babysit","design","pptx","0-1-0-ff47fdd7:ulw-loop","0-1-0-ff47fdd7:comment-checker","best-of-n","xlsx","lina-decision-protocol","implement","0-1-0-ff47fdd7:programming","0-1-0-ff47fdd7:init-deep","0-1-0-ff47fdd7:lsp","mfds-food-safety","0-1-0-ff47fdd7:ulw-plan","0-1-0-ff47fdd7:frontend-ui-ux","0-1-0-ff47fdd7:refactor","execute-plan","review","check-work","docx"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bc8-7d42-8624-2db112419178/updates.jsonl:1:{"timestamp":1780386786,"method":"session/update","params":{"sessionId":"019e8752-6bc8-7d42-8624-2db112419178","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4256,"eventId":"019e8752-6bc8-7d42-8624-2db112419178-0","agentTimestampMs":1780386786976,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8752-6bc8-7d42-8624-2db112419178/updates.jsonl:3:{"timestamp":1780386787,"method":"session/update","params":{"sessionId":"019e8752-6bc8-7d42-8624-2db112419178","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e8752-6bc8-7d42-8624-2db112419178-1","agentTimestampMs":1780386786979,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8344-d0f5-7821-8d05-b2b9f53332e2/updates.jsonl:1:{"timestamp":1780318786,"method":"session/update","params":{"sessionId":"019e8344-d0f5-7821-8d05-b2b9f53332e2","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"0-1-0-ff47fdd7:start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"0-1-0-ff47fdd7:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"0-1-0-ff47fdd7:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}},{"name":"cancel","description":"Cancel any active LFG mode (ulw, ralph, team, atlas, etc.) and record the cancellation state.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/cancel/SKILL.md"}},{"name":"agent-browser","description":"Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/agent-browser/SKILL.md"}},{"name":"pre-publish-review","description":"Nuclear-grade 16-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (5 agents) for holistic review, and 1 oracle for overall release synthesis. Use before EVERY npm publish. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/pre-publish-review/SKILL.md"}},{"name":"remove-deadcode","description":"Remove unused code from this project with ultrawork mode, LSP-verified safety, atomic commits. Triggers: remove dead code, dead code, cleanup, remove unused.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/remove-deadcode/SKILL.md"}},{"name":"github-triage","description":"Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/github-triage/SKILL.md"}},{"name":"ai-slop-remover","description":"Removes AI-generated code smells from a SINGLE file while preserving functionality. For multiple files, call in PARALLEL per file.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ai-slop-remover/SKILL.md"}},{"name":"plan","description":"Strategic planning with Prometheus: Interview-mode planning that produces verifiable plans.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/plan/SKILL.md"}},{"name":"work-with-pr","description":"Full PR lifecycle: git worktree → implement → atomic commits → PR creation → verification loop (CI + review-work + Cubic approval) → merge. Keeps iterating until ALL gates pass and PR is merged. Worktree auto-cleanup after merge. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/work-with-pr/SKILL.md"}},{"name":"team-mode","description":"Team orchestration — create and manage parallel agent teams (OFF by default; enable via team_mode.enabled in config). Loading this skill provides usage documentation; the team_* tools are registered globally when team_mode.enabled=true and access-gated by team role.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team-mode/SKILL.md"}},{"name":"doctor","description":"Diagnose and fix LFG plugin state, schema, and provider configuration issues.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/doctor/SKILL.md"}},{"name":"get-unpublished-changes","description":"Compare HEAD with the latest published npm version and list all unpublished changes. Triggers: unpublished changes, changelog, what changed, whats new.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/get-unpublished-changes/SKILL.md"}},{"name":"setup","description":"Install/sync the LFG Grok plugin and configure provider/model metadata.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/setup/SKILL.md"}},{"name":"provider","description":"List, show, or add provider/model metadata without storing secret values.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/provider/SKILL.md"}},{"name":"ralph","description":"Durable self-referential development loop that continues until the objective is met.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ralph/SKILL.md"}},{"name":"omomomo","description":"Easter egg command - about oh-my-opencode. Triggers: omomomo, about, easter egg.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/omomomo/SKILL.md"}},{"name":"playwright","description":"MUST USE for any browser-related tasks. Browser automation via Playwright MCP - verification, browsing, information gathering, web scraping, testing, screenshots, and all browser interactions.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/playwright/SKILL.md"}},{"name":"publish","description":"Publish oh-my-opencode to npm via GitHub Actions workflow. Argument: <patch|minor|major>. Triggers: publish, release, deploy, npm publish.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/publish/SKILL.md"}},{"name":"git-master","description":"MUST USE for ANY git operations. Atomic commits, rebase/squash, history search (blame, bisect, log -S). STRONGLY RECOMMENDED: Use with task(category='quick', load_skills=['git-master'], ...) to save context. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/git-master/SKILL.md"}},{"name":"team","description":"Create and manage durable OMO-style agent teams with tmux-backed local observability.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/team/SKILL.md"}},{"name":"lfg","description":"Use this skill when validating the `islee23520/lfg` plugin package.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/lfg/SKILL.md"}},{"name":"lfg:start-work","description":"Start or resume Atlas execution for the active plan. Routes through Atlas/Boulder semantics.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/start-work/SKILL.md"}},{"name":"hyperplan","description":"Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', 'adversarial plan', 'hostile planning', 'cross-critique plan', '하이퍼플랜', '적대적 계획', '교차 비평'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/hyperplan/SKILL.md"}},{"name":"dev-browser","description":"Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include \"go to [url]\", \"click on\", \"fill out the form\", \"take a screenshot\", \"scrape\", \"automate\", \"test the website\", \"log into\", or any browser interaction request.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/dev-browser/SKILL.md"}},{"name":"worker","description":"Sisyphus-Junior category dispatch and team worker protocol (ACK, mailbox, task lifecycle).","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/worker/SKILL.md"}},{"name":"ulw","description":"Activate OMO-style Ultrawork mode: Sisyphus-led autonomous execution with Boulder persistence.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/ulw/SKILL.md"}},{"name":"lfg:review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/review-work/SKILL.md"}},{"name":"lfg:frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/plugins/lfg/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","spawn_subagent","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4426,"eventId":"019e8344-d0f5-7821-8d05-b2b9f53332e2-0","agentTimestampMs":1780318786331,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":71}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8344-d0f5-7821-8d05-b2b9f53332e2/updates.jsonl:3:{"timestamp":1780318813,"method":"_x.ai/session/update","params":{"sessionId":"019e8344-d0f5-7821-8d05-b2b9f53332e2","update":{"sessionUpdate":"hooks_changed","hooks":[{"name":"global/cmux-session:session_start[0].hooks[0]","event":"session_start","handlerType":"command","matcher":null,"command":": cmux-grok-hook-v2; :; printenv CMUX_GROK_HOOKS_DISABLED | grep -qx 1 && { :; echo '{}'; } || { if [ -x '/Applications/cmux.app/Contents/Resources/bin/cmux' ]; then '/Applications/cmux.app/Contents/Resources/bin/cmux' --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok session-start; elif command -v cmux >/dev/null 2>&1; then cmux --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok session-start; else echo '{}'; fi; cmux_hook_status=$?; :; exit $cmux_hook_status; }","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:session_start[0].hooks[0]","event":"session_start","handlerType":"command","matcher":null,"command":"node \"${PLUGIN_ROOT}/components/rules/dist/cli.js\" hook session-start","url":null,"timeoutMs":10000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:session_start[1].hooks[0]","event":"session_start","handlerType":"command","matcher":null,"command":"node \"${PLUGIN_ROOT}/components/telemetry/dist/cli.js\" hook session-start","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/lfg/hooks:session_start[0].hooks[0]","event":"session_start","handlerType":"command","matcher":null,"command":"scripts/lfg-audit-hook.ts","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/plugins/lfg/hooks","disabled":false},{"name":"global/cmux-session:user_prompt_submit[0].hooks[0]","event":"user_prompt_submit","handlerType":"command","matcher":null,"command":": cmux-grok-hook-v2; :; printenv CMUX_GROK_HOOKS_DISABLED | grep -qx 1 && { :; echo '{}'; } || { if [ -x '/Applications/cmux.app/Contents/Resources/bin/cmux' ]; then '/Applications/cmux.app/Contents/Resources/bin/cmux' --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok prompt-submit; elif command -v cmux >/dev/null 2>&1; then cmux --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok prompt-submit; else echo '{}'; fi; cmux_hook_status=$?; :; exit $cmux_hook_status; }","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:user_prompt_submit[0].hooks[0]","event":"user_prompt_submit","handlerType":"command","matcher":null,"command":"node \"${PLUGIN_ROOT}/components/rules/dist/cli.js\" hook user-prompt-submit","url":null,"timeoutMs":10000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:user_prompt_submit[1].hooks[0]","event":"user_prompt_submit","handlerType":"command","matcher":null,"command":"node \"${PLUGIN_ROOT}/components/ultrawork/dist/cli.js\" hook user-prompt-submit","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:user_prompt_submit[2].hooks[0]","event":"user_prompt_submit","handlerType":"command","matcher":null,"command":"node \"${PLUGIN_ROOT}/components/ulw-loop/dist/cli.js\" hook user-prompt-submit","url":null,"timeoutMs":10000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/lfg/hooks:user_prompt_submit[0].hooks[0]","event":"user_prompt_submit","handlerType":"command","matcher":null,"command":"scripts/lfg-goal-harness.ts","url":null,"timeoutMs":8000,"sourceDir":"/Users/ilseoblee/.grok/plugins/lfg/hooks","disabled":false},{"name":"global/cmux-session:pre_tool_use[0].hooks[0]","event":"pre_tool_use","handlerType":"command","matcher":null,"command":": cmux-grok-hook-v2; :; printenv CMUX_GROK_HOOKS_DISABLED | grep -qx 1 && { :; echo '{}'; } || { if [ -x '/Applications/cmux.app/Contents/Resources/bin/cmux' ]; then '/Applications/cmux.app/Contents/Resources/bin/cmux' --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks feed --source grok --event PreToolUse; elif command -v cmux >/dev/null 2>&1; then cmux --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks feed --source grok --event PreToolUse; else echo '{}'; fi; cmux_hook_status=$?; :; exit $cmux_hook_status; }","url":null,"timeoutMs":120000,"sourceDir":"/Users/ilseoblee/.grok/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:pre_tool_use[0].hooks[0]","event":"pre_tool_use","handlerType":"command","matcher":"^create_goal$","command":"node \"${PLUGIN_ROOT}/components/ulw-loop/dist/cli.js\" hook pre-tool-use","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/lfg/hooks:pre_tool_use[0].hooks[0]","event":"pre_tool_use","handlerType":"command","matcher":null,"command":"scripts/lfg-audit-hook.ts","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/plugins/lfg/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:post_tool_use[0].hooks[0]","event":"post_tool_use","handlerType":"command","matcher":"^(apply_patch|write|Write|edit|Edit|multi_edit|multiedit|MultiEdit)$","command":"node \"${PLUGIN_ROOT}/components/comment-checker/dist/cli.js\" hook post-tool-use","url":null,"timeoutMs":30000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:post_tool_use[0].hooks[1]","event":"post_tool_use","handlerType":"command","matcher":"^(apply_patch|write|Write|edit|Edit|multi_edit|multiedit|MultiEdit)$","command":"node \"${PLUGIN_ROOT}/components/lsp/dist/cli.js\" hook post-tool-use","url":null,"timeoutMs":60000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:post_tool_use[1].hooks[0]","event":"post_tool_use","handlerType":"command","matcher":"^apply_patch$","command":"node \"${PLUGIN_ROOT}/components/rules/dist/cli.js\" hook post-tool-use","url":null,"timeoutMs":10000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/lfg/hooks:post_tool_use[0].hooks[0]","event":"post_tool_use","handlerType":"command","matcher":null,"command":"scripts/lfg-goal-harness.ts","url":null,"timeoutMs":8000,"sourceDir":"/Users/ilseoblee/.grok/plugins/lfg/hooks","disabled":false},{"name":"global/cmux-session:stop[0].hooks[0]","event":"stop","handlerType":"command","matcher":null,"command":": cmux-grok-hook-v2; :; printenv CMUX_GROK_HOOKS_DISABLED | grep -qx 1 && { :; echo '{}'; } || { if [ -x '/Applications/cmux.app/Contents/Resources/bin/cmux' ]; then '/Applications/cmux.app/Contents/Resources/bin/cmux' --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok stop; elif command -v cmux >/dev/null 2>&1; then cmux --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok stop; else echo '{}'; fi; cmux_hook_status=$?; :; exit $cmux_hook_status; }","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/hooks","disabled":false},{"name":"plugin/0-1-0-ff47fdd7/hooks:stop[0].hooks[0]","event":"stop","handlerType":"command","matcher":null,"command":"node \"${PLUGIN_ROOT}/components/start-work-continuation/dist/cli.js\" hook stop","url":null,"timeoutMs":10000,"sourceDir":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/hooks","disabled":false},{"name":"plugin/lfg/hooks:stop[0].hooks[0]","event":"stop","handlerType":"command","matcher":null,"command":"scripts/lfg-goal-harness.ts","url":null,"timeoutMs":8000,"sourceDir":"/Users/ilseoblee/.grok/plugins/lfg/hooks","disabled":false},{"name":"global/cmux-session:notification[0].hooks[0]","event":"notification","handlerType":"command","matcher":null,"command":": cmux-grok-hook-v2; :; printenv CMUX_GROK_HOOKS_DISABLED | grep -qx 1 && { :; echo '{}'; } || { if [ -x '/Applications/cmux.app/Contents/Resources/bin/cmux' ]; then '/Applications/cmux.app/Contents/Resources/bin/cmux' --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok notification; elif command -v cmux >/dev/null 2>&1; then cmux --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok notification; else echo '{}'; fi; cmux_hook_status=$?; :; exit $cmux_hook_status; }","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/hooks","disabled":false},{"name":"plugin/lfg/hooks:notification[0].hooks[0]","event":"notification","handlerType":"command","matcher":null,"command":"scripts/lfg-audit-hook.ts","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/plugins/lfg/hooks","disabled":false},{"name":"global/cmux-session:session_end[0].hooks[0]","event":"session_end","handlerType":"command","matcher":null,"command":": cmux-grok-hook-v2; :; printenv CMUX_GROK_HOOKS_DISABLED | grep -qx 1 && { :; echo '{}'; } || { if [ -x '/Applications/cmux.app/Contents/Resources/bin/cmux' ]; then '/Applications/cmux.app/Contents/Resources/bin/cmux' --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok session-end; elif command -v cmux >/dev/null 2>&1; then cmux --socket '/Users/ilseoblee/Library/Application Support/cmux/cmux.sock' hooks grok session-end; else echo '{}'; fi; cmux_hook_status=$?; :; exit $cmux_hook_status; }","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/hooks","disabled":false},{"name":"plugin/lfg/hooks:session_end[0].hooks[0]","event":"session_end","handlerType":"command","matcher":null,"command":"scripts/lfg-audit-hook.ts","url":null,"timeoutMs":5000,"sourceDir":"/Users/ilseoblee/.grok/plugins/lfg/hooks","disabled":false}],"project_trusted":false}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8344-d0f5-7821-8d05-b2b9f53332e2/updates.jsonl:4:{"timestamp":1780318813,"method":"_x.ai/session/update","params":{"sessionId":"019e8344-d0f5-7821-8d05-b2b9f53332e2","update":{"sessionUpdate":"plugins_changed","plugins":[{"name":"0-1-0-ff47fdd7","id":"user/631e7a62/0-1-0-ff47fdd7","root":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7","scope":"user","trusted":true,"enabled":true,"version":null,"description":null,"skillCount":13,"skillNames":["init-deep","ulw-loop","debugging","lsp","comment-checker","programming","remove-ai-slops","start-work","rules","ulw-plan","review-work","refactor","frontend-ui-ux"],"agentCount":0,"hookStatus":"active","hookCount":12,"mcpServerCount":2,"mcpStatus":"active"},{"name":"agent-sdk-dev","id":"user/9c5c2f39/agent-sdk-dev","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/agent-sdk-dev","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Claude Agent SDK Development Plugin","skillCount":1,"skillNames":["new-sdk-app"],"agentCount":2,"agentNames":["agent-sdk-verifier-ts","agent-sdk-verifier-py"],"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"asana","id":"user/1a302aeb/asana","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/asana","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Asana project management integration. Create and manage tasks, search projects, update assignments, track progress, and integrate your development workflow with Asana's work management platform.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"claude-code-setup","id":"user/ad4eff69/claude-code-setup","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/claude-code-setup","scope":"user","trusted":true,"enabled":false,"version":"1.0.0","description":"Analyze codebases and recommend tailored Claude Code automations such as hooks, skills, MCP servers, and subagents.","skillCount":1,"skillNames":["claude-automation-recommender"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"claude-md-management","id":"user/9bc94cfd/claude-md-management","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/claude-md-management","scope":"user","trusted":true,"enabled":false,"version":"1.0.0","description":"Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.","skillCount":2,"skillNames":["claude-md-improver","revise-claude-md"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"code-modernization","id":"user/23b3688c/code-modernization","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/code-modernization","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Modernize legacy codebases (COBOL, legacy Java/C++, monolith web apps) with a structured assess → map → extract-rules → reimagine → transform → harden workflow and specialist review agents","skillCount":7,"skillNames":["modernize-extract-rules","modernize-transform","modernize-map","modernize-reimagine","modernize-assess","modernize-harden","modernize-brief"],"agentCount":5,"agentNames":["architecture-critic","test-engineer","security-auditor","legacy-analyst","business-rules-extractor"],"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"code-review","id":"user/14b9e639/code-review","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/code-review","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Automated code review for pull requests using multiple specialized agents with confidence-based scoring","skillCount":1,"skillNames":["code-review"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"code-simplifier","id":"user/8d49e924/code-simplifier","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/code-simplifier","scope":"user","trusted":true,"enabled":false,"version":"1.0.0","description":"Agent that simplifies and refines code for clarity, consistency, and maintainability while preserving functionality","skillCount":0,"agentCount":1,"agentNames":["code-simplifier"],"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"commit-commands","id":"user/9aecd669/commit-commands","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/commit-commands","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Streamline your git workflow with simple commands for committing, pushing, and creating pull requests","skillCount":3,"skillNames":["commit-push-pr","clean_gone","commit"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"context7","id":"user/b02843c0/context7","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/context7","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"discord","id":"user/8db254a7/discord","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/discord","scope":"user","trusted":true,"enabled":false,"version":"0.0.4","description":"Discord channel for Claude Code — messaging bridge with built-in access control. Manage pairing, allowlists, and policy via /discord:access.","skillCount":2,"skillNames":["access","configure"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":1,"mcpStatus":"active"},{"name":"example-plugin","id":"user/6c789d9d/example-plugin","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/example-plugin","scope":"user","trusted":true,"enabled":false,"version":null,"description":"A comprehensive example plugin demonstrating all Claude Code extension options including commands, agents, skills, hooks, and MCP servers","skillCount":3,"skillNames":["example-skill","example-command","example-command"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"explanatory-output-style","id":"user/ff737ff7/explanatory-output-style","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/explanatory-output-style","scope":"user","trusted":true,"enabled":false,"version":"1.0.0","description":"Adds educational insights about implementation choices and codebase patterns (mimics the deprecated Explanatory output style)","skillCount":0,"agentCount":0,"hookStatus":"active","hookCount":1,"mcpServerCount":0,"mcpStatus":"none"},{"name":"fakechat","id":"user/fd618ef6/fakechat","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/fakechat","scope":"user","trusted":true,"enabled":false,"version":"0.0.1","description":"Localhost iMessage-style web chat for Claude Code — test surface with file upload and edits. No tokens, no access control.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":1,"mcpStatus":"active"},{"name":"feature-dev","id":"user/488466ff/feature-dev","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/feature-dev","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review","skillCount":1,"skillNames":["feature-dev"],"agentCount":3,"agentNames":["code-reviewer","code-explorer","code-architect"],"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"firebase","id":"user/401410cd/firebase","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/firebase","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Google Firebase MCP integration. Manage Firestore databases, authentication, cloud functions, hosting, and storage. Build and manage your Firebase backend directly from your development workflow.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"frontend-design","id":"user/8fa6766e/frontend-design","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/frontend-design","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Frontend design skill for UI/UX implementation","skillCount":1,"skillNames":["frontend-design"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"github","id":"user/0acb3724/github","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/github","scope":"user","trusted":true,"enabled":true,"version":null,"description":"Official GitHub MCP server for repository management. Create issues, manage pull requests, review code, search repositories, and interact with GitHub's full API directly from Claude Code.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none","conflict":"Name collision: shadowing \"github\" from /Users/ilseoblee/.claude/plugins/cache/claude-plugins-official/github/unknown"},{"name":"gitlab","id":"user/6e2981c8/gitlab","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/gitlab","scope":"user","trusted":true,"enabled":false,"version":null,"description":"GitLab DevOps platform integration. Manage repositories, merge requests, CI/CD pipelines, issues, and wikis. Full access to GitLab's comprehensive DevOps lifecycle tools.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"greptile","id":"user/899a3ddd/greptile","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/greptile","scope":"user","trusted":true,"enabled":false,"version":null,"description":"AI code review agent for GitHub and GitLab. View and resolve Greptile's PR review comments directly from Claude Code.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"hookify","id":"user/7f38c745/hookify","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/hookify","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Easily create hooks to prevent unwanted behaviors by analyzing conversation patterns","skillCount":5,"skillNames":["writing-rules","help","list","configure","hookify"],"agentCount":1,"agentNames":["conversation-analyzer"],"hookStatus":"active","hookCount":4,"mcpServerCount":0,"mcpStatus":"none"},{"name":"imessage","id":"user/bc96e2ec/imessage","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/imessage","scope":"user","trusted":true,"enabled":false,"version":"0.1.0","description":"iMessage channel for Claude Code — reads chat.db directly, sends via AppleScript. Built-in access control; manage pairing, allowlists, and policy via /imessage:access.","skillCount":2,"skillNames":["access","configure"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":1,"mcpStatus":"active"},{"name":"laravel-boost","id":"user/2d759089/laravel-boost","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/laravel-boost","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Laravel development toolkit MCP server. Provides intelligent assistance for Laravel applications including Artisan commands, Eloquent queries, routing, migrations, and framework-specific code generation.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"learning-output-style","id":"user/8dd7b61c/learning-output-style","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/learning-output-style","scope":"user","trusted":true,"enabled":false,"version":"1.0.0","description":"Interactive learning mode that requests meaningful code contributions at decision points (mimics the unshipped Learning output style)","skillCount":0,"agentCount":0,"hookStatus":"active","hookCount":1,"mcpServerCount":0,"mcpStatus":"none"},{"name":"lfg","id":"user/c195d9e5/lfg","root":"/Users/ilseoblee/.grok/plugins/lfg","scope":"user","trusted":true,"enabled":true,"version":"0.0.1","description":"OMO agent hierarchy parity runtime for Grok Build.","skillCount":27,"skillNames":["cancel","agent-browser","pre-publish-review","remove-deadcode","github-triage","ai-slop-remover","plan","work-with-pr","team-mode","doctor","get-unpublished-changes","setup","provider","ralph","omomomo","playwright","publish","git-master","team","lfg","start-work","hyperplan","dev-browser","worker","ulw","review-work","frontend-ui-ux"],"agentCount":12,"agentNames":["prometheus","sisyphus","explore","multimodal-looker","sisyphus-junior","momus","metis","oracle","atlas","builtin-agents","hephaestus","librarian"],"hookStatus":"active","hookCount":9,"mcpServerCount":1,"mcpStatus":"active"},{"name":"linear","id":"user/07c7b589/linear","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/linear","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Linear issue tracking integration. Create issues, manage projects, update statuses, search across workspaces, and streamline your software development workflow with Linear's modern issue tracker.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"math-olympiad","id":"user/56062eb1/math-olympiad","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/math-olympiad","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Solve competition math (IMO, Putnam, USAMO) with adversarial verification that catches what self-verification misses. Fresh-context verifiers attack proofs with specific failure patterns. Calibrated abstention over bluffing.","skillCount":1,"skillNames":["math-olympiad"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"mcp-server-dev","id":"user/dc3c3339/mcp-server-dev","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/mcp-server-dev","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Skills for designing and building MCP servers that work seamlessly with Claude — guides you through deployment models (remote HTTP, MCPB, local), tool design patterns, auth, and interactive MCP apps.","skillCount":3,"skillNames":["build-mcp-server","build-mcp-app","build-mcpb"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"playground","id":"user/6d1563fd/playground","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/playground","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Creates interactive HTML playgrounds — self-contained single-file explorers with visual controls, live preview, and prompt output with copy button","skillCount":1,"skillNames":["playground"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"playwright","id":"user/b9ee4843/playwright","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/playwright","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Browser automation and end-to-end testing MCP server by Microsoft. Enables Claude to interact with web pages, take screenshots, fill forms, click elements, and perform automated browser testing workflows.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"plugin-dev","id":"user/24dbce98/plugin-dev","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/plugin-dev","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Plugin development toolkit with skills for creating agents, commands, hooks, MCP integrations, and comprehensive plugin structure guidance","skillCount":8,"skillNames":["command-development","skill-development","plugin-settings","plugin-structure","hook-development","mcp-integration","agent-development","create-plugin"],"agentCount":3,"agentNames":["agent-creator","skill-reviewer","plugin-validator"],"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"pr-review-toolkit","id":"user/f9238f8f/pr-review-toolkit","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/pr-review-toolkit","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification","skillCount":1,"skillNames":["review-pr"],"agentCount":6,"agentNames":["code-reviewer","silent-failure-hunter","code-simplifier","comment-analyzer","pr-test-analyzer","type-design-analyzer"],"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"ralph-loop","id":"user/816754b1/ralph-loop","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/ralph-loop","scope":"user","trusted":true,"enabled":false,"version":"1.0.0","description":"Continuous self-referential AI loops for interactive iterative development, implementing the Ralph Wiggum technique. Run Claude in a while-true loop with the same prompt until task completion.","skillCount":3,"skillNames":["help","cancel-ralph","ralph-loop"],"agentCount":0,"hookStatus":"active","hookCount":1,"mcpServerCount":0,"mcpStatus":"none"},{"name":"security-guidance","id":"user/b1ae1dfa/security-guidance","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/security-guidance","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Security reminder hook that warns about potential security issues when editing files, including command injection, XSS, and unsafe code patterns","skillCount":0,"agentCount":0,"hookStatus":"active","hookCount":1,"mcpServerCount":0,"mcpStatus":"none"},{"name":"serena","id":"user/cdfded47/serena","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/serena","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Semantic code analysis MCP server providing intelligent code understanding, refactoring suggestions, and codebase navigation through language server protocol integration.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"session-report","id":"user/cdf50829/session-report","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/session-report","scope":"user","trusted":true,"enabled":false,"version":null,"description":null,"skillCount":1,"skillNames":["session-report"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"skill-creator","id":"user/8baf8176/skill-creator","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/plugins/skill-creator","scope":"user","trusted":true,"enabled":false,"version":null,"description":"Create new skills, improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, or benchmark skill performance with variance analysis.","skillCount":1,"skillNames":["skill-creator"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"},{"name":"telegram","id":"user/1da31ab5/telegram","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/telegram","scope":"user","trusted":true,"enabled":false,"version":"0.0.6","description":"Telegram channel for Claude Code — messaging bridge with built-in access control. Manage pairing, allowlists, and policy via /telegram:access.","skillCount":2,"skillNames":["access","configure"],"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":1,"mcpStatus":"active"},{"name":"terraform","id":"user/39d44f5d/terraform","root":"/Users/ilseoblee/.claude/plugins/marketplaces/claude-plugins-official/external_plugins/terraform","scope":"user","trusted":true,"enabled":false,"version":null,"description":"The Terraform MCP Server provides seamless integration with Terraform ecosystem, enabling advanced automation and interaction capabilities for Infrastructure as Code (IaC) development.","skillCount":0,"agentCount":0,"hookStatus":"none","hookCount":0,"mcpServerCount":0,"mcpStatus":"none"}]}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3f94-72b1-8fb7-213fa40bc89f/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3f94-72b1-8fb7-213fa40bc89f/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["cua-driver","design","k-skill-setup","0-1-0-ff47fdd7:init-deep","0-1-0-ff47fdd7:refactor","0-1-0-ff47fdd7:debugging","check-work","korean-law-search","0-1-0-ff47fdd7:start-work","0-1-0-ff47fdd7:programming","implement","0-1-0-ff47fdd7:frontend-ui-ux","github-yeet","xlsx","mfds-food-safety","0-1-0-ff47fdd7:comment-checker","0-1-0-ff47fdd7:remove-ai-slops","execute-plan","pr-babysit","pptx","0-1-0-ff47fdd7:lsp","0-1-0-ff47fdd7:ulw-plan","lina-decision-protocol","0-1-0-ff47fdd7:ulw-loop","create-skill","help","review","0-1-0-ff47fdd7:rules","docx","0-1-0-ff47fdd7:review-work","best-of-n"]}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3f94-72b1-8fb7-213fa40bc89f/updates.jsonl:1:{"timestamp":1780377076,"method":"session/update","params":{"sessionId":"019e86be-3f94-72b1-8fb7-213fa40bc89f","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4195,"eventId":"019e86be-3f94-72b1-8fb7-213fa40bc89f-0","agentTimestampMs":1780377076279,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e86be-3f94-72b1-8fb7-213fa40bc89f/updates.jsonl:3:{"timestamp":1780377077,"method":"session/update","params":{"sessionId":"019e86be-3f94-72b1-8fb7-213fa40bc89f","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e86be-3f94-72b1-8fb7-213fa40bc89f-1","agentTimestampMs":1780377076280,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e7466-d18d-71b0-9b1f-6599dc5defbc/terminal/call-8c072a29-7920-4aaa-82f5-613f9dfaf314-140.log:18:- Scope limit: only model/provider surface, or also a minimal "Codex host session adapter" in LFG hooks for when Grok is invoked from a LazyCodex/Codex environment? (Rec: model surface only for this ticket; second slice if user confirms.)
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8754-cf41-7892-90a4-9517df504294/updates.jsonl:1:{"timestamp":1780386943,"method":"session/update","params":{"sessionId":"019e8754-cf41-7892-90a4-9517df504294","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":4256,"eventId":"019e8754-cf41-7892-90a4-9517df504294-0","agentTimestampMs":1780386943362,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8754-cf41-7892-90a4-9517df504294/updates.jsonl:3:{"timestamp":1780386944,"method":"session/update","params":{"sessionId":"019e8754-cf41-7892-90a4-9517df504294","update":{"sessionUpdate":"available_commands_update","availableCommands":[{"name":"compact","description":"Compress conversation history to save context window","input":{"hint":"optional context about what to preserve"}},{"name":"always-approve","description":"Toggle always-approve mode (skip all permission prompts)","input":{"hint":"on|off"}},{"name":"context","description":"Show context window usage and session stats","input":null},{"name":"hooks-trust","description":"Trust this project for hook execution","input":null},{"name":"hooks-list","description":"Show hooks loaded in this session","input":null},{"name":"hooks-add","description":"Add a custom hook file or directory","input":{"hint":"path to hook file or directory"}},{"name":"hooks-remove","description":"Remove a custom hook file or directory path","input":{"hint":"path to hook file or directory"}},{"name":"hooks-untrust","description":"Remove trust for the current project","input":null},{"name":"plugins","description":"Manage plugins (list, reload, trust, add, remove)","input":{"hint":"list | reload | trust <path> | add <path> | remove <path>"}},{"name":"reload-plugins","description":"Reload plugins from disk (alias for /plugins reload)","input":null},{"name":"session-info","description":"Show session details (model, turns, context usage)","input":null},{"name":"feedback","description":"Send feedback about the current session","input":{"hint":"feedback text"}},{"name":"loop","description":"Run a prompt on a recurring interval","input":{"hint":"[interval] <prompt>"}},{"name":"xlsx","description":"Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like \"the xlsx in my downloads\") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/xlsx/SKILL.md"}},{"name":"best-of-n","description":"Parallel implementation tournament","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/best-of-n/SKILL.md"}},{"name":"create-skill","description":"Create a new Grok skill","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/create-skill/SKILL.md"}},{"name":"pptx","description":"Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions \"deck,\" \"slides,\" \"presentation,\" or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/pptx/SKILL.md"}},{"name":"check-work","description":"Verify changes with a subagent","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/check-work/SKILL.md"}},{"name":"docx","description":"Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/docx/SKILL.md"}},{"name":"help","description":"Grok docs — config, MCP, auth, skills, commands","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/skills/help/SKILL.md"}},{"name":"mfds-food-safety","description":"식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md"}},{"name":"cua-driver","description":"Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/cua-driver/SKILL.md"}},{"name":"k-skill-setup","description":"After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md"}},{"name":"korean-law-search","description":"Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md"}},{"name":"github-yeet","description":"Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages. Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/github-yeet/SKILL.md"}},{"name":"lina-decision-protocol","description":"Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md"}},{"name":"design","description":"Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.","input":{"hint":"<description of what to design>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/design/SKILL.md"}},{"name":"execute-plan","description":"Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.","input":{"hint":"<design-doc-path> [--effort N] [--concurrency N] [--dry-run] [--resume <PLAN_ID>] [--instructions \"...\"] [--no-graphite] [--auto-pr]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md"}},{"name":"pr-babysit","description":"Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).","input":{"hint":"add <number> | remove <number> | list | check"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md"}},{"name":"review","description":"Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.","input":{"hint":"[--local | --branch <name> | --pr <number-or-url> | <auto-detect>]"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/review/SKILL.md"}},{"name":"implement","description":"Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.","input":{"hint":"[--effort N] <description of what to implement>"},"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md"}},{"name":"init-deep","description":"(builtin) Initialize hierarchical AGENTS.md knowledge base","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md"}},{"name":"ulw-loop","description":"Goal-like ultrawork loop for systematic decomposition","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md"}},{"name":"debugging","description":"MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix minimally, QA by actually USING the system, scrub artifacts. The actual HOW lives in `references/` — READ THEM. Triggers: 'debug this', 'why is X not working', 'hanging', 'attach a debugger', 'reverse engineer', 'pwndbg', 'gdb', 'lldb', 'node inspect', 'tsx debug', 'pdb', 'dlv', 'delve', 'rust-gdb', 'set a breakpoint', 'context window exploded', 'why is the response empty', 'attach the debugger', 'debug it', 'why is this happening', 'trace this bug', 'reproduce and fix', 'silent failure', 'HTTP 200 but empty', 'why did it stop', 'inspect the binary', 'reverse engineering', 'playwright'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md"}},{"name":"lsp","description":"Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md"}},{"name":"comment-checker","description":"Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md"}},{"name":"programming","description":"MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes to references/{python,rust,typescript,rust-ub,go}/. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md"}},{"name":"remove-ai-slops","description":"Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pure LOC with mandatory modular refactoring). MUST USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md"}},{"name":"start-work","description":"Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md"}},{"name":"rules","description":"Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md"}},{"name":"ulw-plan","description":"Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md"}},{"name":"review-work","description":"Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'review my work', 'review changes', 'QA my work', 'verify implementation', 'check my work', 'validate changes', 'post-implementation review'.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md"}},{"name":"refactor","description":"Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md"}},{"name":"frontend-ui-ux","description":"Designer-turned-developer who crafts stunning UI/UX even without design mockups","input":null,"_meta":{"scope":"user","path":"/Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md"}}],"_meta":{"tools":["run_terminal_command","read_file","search_replace","list_dir","grep","kill_command_or_subagent","todo_write","get_command_or_subagent_output","wait_commands_or_subagents","scheduler_create","scheduler_delete","scheduler_list","monitor","search_tool","use_tool","update_goal","enter_plan_mode","exit_plan_mode","ask_user_question","web_search","image_gen","image_edit","video_gen","write"]}},"_meta":{"totalTokens":6006,"eventId":"019e8754-cf41-7892-90a4-9517df504294-1","agentTimestampMs":1780386943363,"updateType":"AvailableCommandsUpdate","updateParams":{"commandsCount":44}}}}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8754-cf41-7892-90a4-9517df504294/chat_history.jsonl:4:{"type":"user","content":[{"type":"text","text":"<system-reminder>\nThe following skills are available for use:\n\n- xlsx: Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger espec…\n  Absolute path: /Users/ilseoblee/.grok/skills/xlsx/SKILL.md\n- best-of-n: Implement a task N ways in parallel and pick the best. Spawns multiple subagents in isolated worktrees, evaluates all candidates, and applies the winner\n  Use when: Use when asked to \"best of n\", \"try multiple approaches\", \"parallel implementations\", \"/best-of-n\", or \"/bon\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/best-of-n/SKILL.md\n- create-skill: Interactively create a new Grok skill (SKILL.md + optional scripts/references)\n  Use when: Use when the user wants to create a skill, scaffold a skill, or runs /create-skill.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/create-skill/SKILL.md\n- pptx: Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; work…\n  Absolute path: /Users/ilseoblee/.grok/skills/pptx/SKILL.md\n- check-work: Check your work with a verification subagent. Spawns a verifier that reviews diffs, runs builds and tests, and evaluates correctness\n  Use when: Use when asked to \"check work\", \"verify changes\", \"self-verify\", \"/check-work\", \"/check\", \"/verify\", or \"/self-verify\".\n\n  Absolute path: /Users/ilseoblee/.grok/skills/check-work/SKILL.md\n- docx: Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing ima…\n  Absolute path: /Users/ilseoblee/.grok/skills/docx/SKILL.md\n- help: Grok documentation and configuration help\n  Use when: Use when users ask about setup, configuration, MCP servers, authentication, skills, slash commands, keyboard shortcuts, or any Grok feature. Also use proactively when you detect a user is having trouble with setup or onboarding.\n\n  Absolute path: /Users/ilseoblee/.grok/skills/help/SKILL.md\n- mfds-food-safety: 식약처/식품안전나라 공개 표면을 k-skill-proxy 경유로 조회하기 전에 증상·섭취상황을 반드시 되묻는 인터뷰형 식품 안전 체크 스킬.\n  Absolute path: /Users/ilseoblee/.agents/skills/mfds-food-safety/SKILL.md\n- cua-driver: Drive a native macOS app via the cua-driver CLI (default) or MCP server — snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot\n  Use when: Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. \"open a file in TextEdit\", \"navigate to /Applications in Finder\", \"click the Save button in Numbers\").\n  Absolute path: /Users/ilseoblee/.agents/skills/cua-driver/SKILL.md\n- k-skill-setup: After installing the full k-skill bundle, configure and verify the shared cross-platform setup, then optionally wire update checks and GitHub starring with explicit user consent.\n  Absolute path: /Users/ilseoblee/.agents/skills/k-skill-setup/SKILL.md\n- korean-law-search: Use korean-law-mcp first for Korean law lookups, and fall back to Beopmang when the primary service is unavailable.\n  Absolute path: /Users/ilseoblee/.agents/skills/korean-law-search/SKILL.md\n- github-yeet: Intelligent git commit workflow. Analyzes staged/unstaged changes, groups them into logical atomic commits, stages selectively, and commits with conventional commit messages\n  Use when: Use when asked to commit, stage, yeet, push changes, or create PRs. Handles staging, committing, and pushing in a single flow.\n  Absolute path: /Users/ilseoblee/.agents/skills/github-yeet/SKILL.md\n- lina-decision-protocol: Lazy entrypoint for Linalab Decision Protocol. Use MCP tools for decision reviews, Socratic intake, ethical business stress tests, and decision history without loading the full LDP corpus.\n  Absolute path: /Users/ilseoblee/.agents/skills/lina-decision-protocol/SKILL.md\n- design: Run the full design-doc-writer and design-doc-reviewer loop until consensus. Produces a polished design document with a PR plan.\n  Use when: Use when asked to \"design\", \"write a design doc\", \"system design\", \"architecture doc\", \"technical spec\", or \"/design\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/design/SKILL.md\n- execute-plan: Execute a PR Plan DAG from a design document. Parses the plan, topologically sorts it, implements PRs in parallel using worktree-isolated subagents, runs mandatory orchestrator-level review, and assembles either a Graphite PR stack or a plain-git branch stack depending on tool availability.\n  Use when: Use when asked to \"execute plan\", \"run the plan\", \"implement the design\", or \"/execute-plan\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/execute-plan/SKILL.md\n- pr-babysit: Monitor PRs, fix CI failures, address review comments, resolve merge conflicts, and restack stacks. Supports independent PRs, Graphite stacks, and GitHub stacked PRs (gh-stack).\n  Use when: Triggers on \"/pr-babysit\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/pr-babysit/SKILL.md\n- review: Run a reviewer subagent against uncommitted local changes, a named branch, or a GitHub PR. Local and branch modes write a review file plus a summary to disk. PR mode posts the findings as a PENDING GitHub review for the user to inspect and submit through the UI.\n  Use when: Use when asked to 'review', 'code review', 'review my changes', 'review this PR', or '/review'.\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/review/SKILL.md\n- implement: Run the full implement-review-fix loop using implementer and reviewer personas. Supports effort-based multi-reviewer scaling (1-5 reviewers) with automatic specialization selection. Includes memory-based feedback loop that learns from past review patterns. Loops until all reviewers find 0 issues of any severity.\n  Use when: Use when asked to \"implement\", \"build\", \"add feature\", \"fix bug\", or \"/implement\".\n  Absolute path: /Users/ilseoblee/.grok/bundled/skills/implement/SKILL.md\n- init-deep: (builtin) Initialize hierarchical AGENTS.md knowledge base\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/init-deep/SKILL.md\n- ulw-loop: Goal-like loop that uses ultrawork mode to decompose work into systematic, evidence-bound steps.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-loop/SKILL.md\n- debugging: MUST USE for any real runtime debugging across ANY language or binary — crashes, silent failures, wrong responses, stuck processes, memory leaks, async misbehavior, unexplained timing, reverse engineering. Runs a hypothesis-driven loop: form ≥3 hypotheses, investigate in parallel, after 2 failed rounds spawn Oracles from orthogonal angles, confirm root cause, lock with a failing test, fix m…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/debugging/SKILL.md\n- lsp: Use when Codex needs language-server diagnostics, definitions, references, symbols, or rename safety checks in the current workspace.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/lsp/SKILL.md\n- comment-checker: Use when Codex needs to understand or respond to automatic comment-checker feedback emitted after an edit-like PostToolUse hook.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/comment-checker/SKILL.md\n- programming: MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD. Routes t…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/programming/SKILL.md\n- remove-ai-slops: Remove AI-generated code smells (slop) from branch changes or an explicit file list. Locks behavior with regression tests FIRST, then runs categorized cleanup via parallel `deep` agents in batches of 5, then verifies with quality gates. Covers 10 slop categories including performance equivalences, excessive complexity (object annotations, if/elif variant chains), and oversized modules (250+ pur…\n  Use when: USE when the user asks to \"remove slop\", \"clean AI code\", \"deslop\", \"clean up AI-generated code\", \"remove AI slop\", or wants to clean up AI-generated patterns from recent changes. Triggers - \"remove ai slops\", \"clean ai code\", \"deslop\", \"cleanup AI generated\", \"remove AI slop\", \"clean up AI-generated code\", \"strip slop\", \"ai-slop cleanup\".\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/remove-ai-slops/SKILL.md\n- start-work: Execute a Prometheus work plan in Codex with Boulder state, evidence ledger updates, worktree discipline, parallel subagents, and Stop-hook continuation. Use after planning when the user says start work, execute plan, continue plan, resume plan, or asks to run a .omo/plans plan.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/start-work/SKILL.md\n- rules: Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or environment configuration.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/rules/SKILL.md\n- ulw-plan: Strategic planning consultant that produces decision-complete work plans through Socratic interview, codebase exploration, Metis gap analysis, and optional Momus high-accuracy review. MUST\n  Use when: USE when the task has 5+ steps, scope is ambiguous, multiple modules are involved, or the user asks for a plan. Triggers: ulw-plan, plan this, create a work plan, interview me, start planning, plan mode, help me plan this, break this down.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/ulw-plan/SKILL.md\n- review-work: Post-implementation review orchestrator. Launches 5 parallel background sub-agents: Oracle (goal/constraint verification), Oracle (code quality), Oracle (security), unspecified-high (hands-on QA execution), unspecified-high (context mining from GitHub/git/Slack/Notion). All must pass for review to pass. MUST USE after completing any significant implementation work. Triggers: 'review work', 'rev…\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/review-work/SKILL.md\n- refactor: Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize.\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/refactor/SKILL.md\n- frontend-ui-ux: Designer-turned-developer who crafts stunning UI/UX even without design mockups\n  Absolute path: /Users/ilseoblee/.grok/installed-plugins/0-1-0-ff47fdd7/skills/frontend-ui-ux/SKILL.md\n</system-reminder>"}],"synthetic_reason":"system_reminder"}
/Users/ilseoblee/.grok/sessions/%2FUsers%2Filseoblee%2Fworkspace%2Flinalab%2Flfg/019e8754-cf41-7892-90a4-9517df504294/announcement_state.json:1:{"mcp_server_fingerprints":{"zai-mcp-server":{"tool_count":8,"description_hash":12161962213042174405,"tool_names_hash":6829172645948100724}},"announced_skill_names":["pptx","github-yeet","0-1-0-ff47fdd7:start-work","0-1-0-ff47fdd7:programming","lina-decision-protocol","review","k-skill-setup","0-1-0-ff47fdd7:comment-checker","0-1-0-ff47fdd7:refactor","check-work","0-1-0-ff47fdd7:debugging","0-1-0-ff47fdd7:rules","0-1-0-ff47fdd7:remove-ai-slops","cua-driver","best-of-n","mfds-food-safety","0-1-0-ff47fdd7:review-work","docx","pr-babysit","0-1-0-ff47fdd7:init-deep","0-1-0-ff47fdd7:ulw-loop","help","create-skill","implement","execute-plan","xlsx","design","korean-law-search","0-1-0-ff47fdd7:ulw-plan","0-1-0-ff47fdd7:lsp","0-1-0-ff47fdd7:frontend-ui-ux"]}
