\n\n## One prompt. A working team.\n\n`dsh-agent-teams` turns the current DeepSeek Harness session into a captain that can assemble durable sub-agents, split a goal into dependency-aware tasks, and coordinate work through direct messages.\n\nAsk in natural language. The plugin provides the team protocol, 14 coordination tools, persistent state, an automatic shared-task scheduler, and a live Web UI—without requiring a separate workflow engine.\n\n
\n \n
\n\n## Releases\n\n[v0.1.22](./release-notes/v0.1.22.md) targets the published Harness `0.2.0-rc.2` packages and retains script-free Git installation. This plugin release uses npm `latest`; the recommended host is Harness `0.2.0-rc.2` (its `next` channel). See the exact [support matrix](./compatibility.json).\n\n## Why AgentTeams?\n\n| Capability | What it changes |\n| --- | --- |\n| **Captain-led delegation** | The current session creates the team, assigns roles, and consolidates the final result. |\n| **Durable members** | Members are continuable DSH sub-agents that can be woken for focused follow-up turns. |\n| **Dependency-aware tasks** | Tasks move through explicit states and cannot be claimed before their dependencies finish. |\n| **Automatic reuse and safe takeover** | Idle members claim the next ready task; reassignment revokes stale attempts before new work starts, and cold recovery retries stranded open attempts. |\n| **Direct messaging** | Members send durable mailbox messages directly to teammates or the captain—no relay required. |\n| **Live activity panel** | The Web UI combines segmented progress, a collapsible roster, and an interactive task DAG; running tasks show the member's model, and completed archives retain their full member and task history. |\n| **Plan before execution** | Normal `/agent-teams` runs stage an unspawned roster and DAG first. The Web panel uses the host model catalog for member routes. Returning to chat stops the planning turn, asks what should change, and revises the same draft; discarding archives the draft, aborts the turn, and explicitly prevents automatic recreation. Only **Approve & Run** enables scheduling; each member starts with its first ready task. |\n| **Quality gates** | Opt-in quality tasks support requirements → implementation → verification → review → integration contracts, automatic repair/re-review, and explicit resume. Scope control is a completion-time audit, not host write interception. See [docs/quality-gates.md](./docs/quality-gates.md). |\n\nThe conversation card and activity panel use Harness's official locale service. They follow live language changes between English and Simplified Chinese—including status labels, dynamic summaries, controls, archive markers, and accessibility text—without a page reload or a separate plugin setting.\n\n## Install and choose versions\n\n**Recommended pair: DeepSeek Harness `0.2.0-rc.2` + AgentTeams `0.1.22`. The host remains an RC.**\n\n| Use case | DeepSeek Harness | AgentTeams plugin |\n| --- | --- | --- |\n| **Recommended** | **`0.2.0-rc.2`** | **`0.1.22`** |\n| Previous recommended RC | `0.1.7-rc.2` | `0.1.22` |\n| Legacy RC | `0.1.5-rc.1`, `0.1.5-rc.2`, `0.1.5-rc.3` | `0.1.22` |\n| Retaining an older RC | `0.1.2-rc.1` | `0.1.22` |\n| Developer Alpha testing | `0.1.2-alpha.5` | `0.1.22` |\n| Retaining an older Alpha | `0.1.2-alpha.2` | `0.1.22` |\n\n### Desktop app\n\nIn the app sidebar, open **Plugins → Add plugin** and enter the npm package spec:\n\n```text\n@nanmicoder/dsh-agent-teams@0.1.22\n```\n\nInstall it, then choose **Enable now**. If the host asks for a restart, restart the desktop app. Use the package spec above for the published build; an npm or GitHub webpage URL is not the same install source. For Git installation, use a commit containing the verified build outputs described below.\n\nThe desktop app supplies its own Harness core and package manager. Upgrading the global CLI does not upgrade that core, and an ordinary standalone CLI cannot manage its desktop profile. The CLI carrier installed by the desktop app can manage that profile in `0.2.0-rc.2`. Check the embedded core against the compatibility list. Harness `0.2.0-rc.2` is now the exact published target; its RC status is unchanged. The earlier source-only `0.2.0` prediction is no longer a support declaration. See the [release migration evidence](./docs/harness-0.2.0-rc.2/README.md).\n\n### CLI / Web: 1. Install DeepSeek Harness\n\n```sh\nnpm install --global @deepseek-ai/dsh@0.2.0-rc.2\ndsh --version\n```\n\nSkip this if you already run this version. Alpha is opt-in: select an exact Alpha version from the table and lock all host dependencies as described in the [maintenance guide](./docs/maintenance-workflow.md).\n\n### CLI / Web: 2. Install the AgentTeams plugin\n\nInstall or upgrade from npm (pinned to this `latest` release):\n\n```sh\ndsh plugin --profile web add --save-exact @nanmicoder/dsh-agent-teams@0.1.22\n```\n\nFor a CLI-managed installation, replace `web` with your active CLI profile. **Stop and restart that profile's Harness process, then refresh the browser.** Installing the plugin does not upgrade Harness; the host and plugin have independent `latest` channels.\n\nFor Git installation, use a Git package spec pinned to a commit containing the verified server, client, and type build outputs in `lib/`, rather than an npm webpage URL. The repository distributes these outputs so installation does not need a `prepare` hook or build-script approval. Source maintainers must run `pnpm build` and update the committed outputs whenever the source changes. Existing npm releases remain immutable.\n\nFor a manually cloned checkout, run `pnpm install --frozen-lockfile` and `pnpm build` after pulling, then restart the corresponding Harness process. Updating Git alone does not update local build output. See the [maintenance guide](./docs/maintenance-workflow.md).\n\n> For older `0.1.0-*` / `0.1.1-*` or unlisted hosts, keep a working pair and follow the [older-version and diagnostic guide](./docs/maintenance-workflow.md).\n\nSee the full [compatibility matrix](./compatibility.json), [source installation and Alpha testing guide](./docs/maintenance-workflow.md), and [verification coverage and platform limits](./docs/maintenance-2026-09-06/release/README.md).\n\nThen ask for a team directly:\n\n> Use AgentTeams to review the commits after v0.5.3 from performance, security, and product perspectives. Return one consolidated report.\n\n## View teams in the workspace\n\nClick **View team** in the current chat header or the team card below a reply to open the native **Team collaboration** tab. The entry belongs to its chat, rather than the general workspace start page.\n\nTeam and task content starts immediately: wide panes use columns and narrow panes stack them. Select a task to locate its owner; member icons open their conversations. Closing the tab does not stop the team, and its chat can reopen it. Completed teams retain their archives. Older hosts keep the activity panel.\n\n## How it works\n\n1. For a request to use AgentTeams, the captain follows the core protocol already in its system instructions. It continues an existing team and uses `agent_teams_status` when current state needs checking. When no team exists, the goal becomes a staged plan for review.\n2. The captain adds role-specific members backed by continuable sub-agents.\n3. The goal becomes tasks with owners and explicit dependencies.\n4. The shared scheduler uses real `running / idle / ready` state to atomically claim one ready task per idle member and wake it. An interrupted resident attempt stays parked and can resume through a direct message without losing its capability; after a cold process restart, the scheduler retries stranded open work with a fresh attempt.\n5. Members update with the current `attempt_id`; reassignment or captain takeover revokes the old attempt and waits for the old worker to quiesce before a new attempt starts.\n6. The captain presents the combined result, then archives the complete team record.\n\nTeam state is stored under `/.agent-teams/`; the Web panel reads that disk truth and combines it with live sub-agent activity.\n\nMember creation is zero-interaction by default: a member on the captain's current LLM route snapshots that provider, model, and reasoning effort, while a member on a requested alternative route snapshots the target model's default effort; later continuations restore the resolved snapshot. Only an explicit heterogeneous-team request (for example, “backend on provider A/model X, frontend on provider B/model Y”) supplies a member-specific `provider` + `model`; there is no per-member model or reasoning prompt.\n\nCaptain sessions keep the concise core protocol and the original 14 native team tools from their first request. All business tools are directly available; no loading tool or extra activation call is needed. Configured profiles retain their bounded directory in the fixed system prompt. Creating, approving, continuing or ending a team does not rewrite the system prompt or tool schemas. Core rules remain available after history compaction or discarded code-mode tool results. Members receive four team tools, fixed member instructions, and their ordinary coding/research tools. Web approval wakes the captain with a control message; later member reports wake it again. See the [fixed protocol and benchmark contract](./docs/progressive-loading.md).\n\n## Slash command\n\nNo “use AgentTeams” phrasing required. The plugin registers the\nclosed-namespace `/agent-teams` host command, so the Web GUI slash menu shows\nan `agent-teams` placeholder with an input hint: pick it (or type the\ncommand), describe the goal, and press Enter.\n\n```\n/agent-teams research the pricing pages of three competitors\n```\n\nThe command pipeline claims the line, then preserves that exact input as an\nordinary user follow-up so it remains visible in the main chat. The gesture\nboundary adds the deterministic activation directive at pre-step, so the\nfirst model request follows the staged planning protocol without a mandatory helper call. The invocation is also durably\nlogged (`command/run` / `command/done`).\n\nSurfaces without command adjudication (for example the headless CLI) get the\nsame deterministic activation through a gesture boundary: any genuine user\nmessage starting with `/agent-teams` activates the protocol for the rest of\nthe text. Mid-sentence mentions stay ordinary prose.\n\nHistorical panels require saved team state or archives. Sessions from early versions that deleted teams without retaining archives do not yet support reconstructing the full panel from logs.\n\n## Configuration\n\nDefaults work without extra setup. A trusted profile can override member behavior:\n\n```yaml\n- id: agent-teams\n config:\n stateDir: .agent-teams\n memberProvider: spawn\n memberModel: deepseek-v4\n memberMaxDepth: 0\n maxMembers: 8\n```\n\n`memberMaxDepth` defaults to `0`: team members cannot create nested subagents. Set `1` to explicitly permit one descendant level; the limit also covers runtime/code-tool calls. Default members report through team messages only, avoiding duplicate native parent reports. Idle roster members make no model requests. Task assignments start distinct turns; coordination joins the nearest model step. Acceptance and consumption are tracked separately. Removal/archive drains the selected branch and its pending input before reporting success.\n\n`memberProvider` is the sub-agent runtime backend (`spawn` / `fork`), not an LLM provider. Cross-LLM-provider routing uses the optional `provider` + `model` fields of `agent_teams_add_member`; `memberModel` is only a model default for all members. A member on the captain's current provider/model inherits the captain's reasoning effort, while a changed provider or model automatically uses the target model's default. To request a particular effort, pass the optional `reasoning_effort` field — one of the target model's supported effort ids, or `\"default\"` to force the model's own default.\n\n`slashCommand: false` disables the deterministic `/agent-teams` activation surfaces (slash command and gesture boundary), leaving the natural-language trigger as the only entry point.\n\n## Boundaries\n\n- One captain leads one active team at a time.\n- Idle members with no open task are automatically reused for ready work. An idle member that still owns an open attempt is parked until messaged or explicitly reassigned; messages that cannot be delivered live remain durable and are retried at a later status boundary.\n- State is file-backed and serialized within one DSH process; concurrent processes editing the same team are not coordinated.\n- The activity panel reports persisted state as-is. Models may occasionally finish work without performing the expected task-state update.\n\nSee [docs/usage.md](./docs/usage.md) for the full tool reference, state model, Web UI behavior, configuration, and known limits.\n\n## Plugin development Skill\n\nCommunity upgrade, audit, benchmark, testing and release skills are vendored with a pinned source revision. See [skills/README.md](./skills/README.md) for local rules and [CONTRIBUTING.md](./CONTRIBUTING.md) for the contribution workflow.\n\nThe repository also ships the open Agent Skills package [`dsh-plugin-development`](./skills/dsh-plugin-development/SKILL.md):\n\n```sh\nnpx skills add NanmiCoder/dsh-agent-teams --skill dsh-plugin-development\n```\n\n## Documentation\n\n| Guide | Covers |\n| --- | --- |\n| [Usage](./docs/usage.md) | Architecture, UI behavior, tools, configuration, limits, and validation |\n| [Verification](./docs/verification-guide.md) | Offline, composition, real e2e, and GUI verification |\n| [Plugin development](./docs/developing-dsh-plugins.md) | Human-readable guide built from this plugin |\n| [README writing](./docs/readme-writing-guide.md) | Repository documentation conventions |\n\n## Development\n\n```sh\npnpm install\npnpm build\npnpm verify\n```\n\n## Named multi-role profiles\n\nConfigure one or more complete team profiles in `cordis.patch.yml`. A profile always supplies the roster (independent provider/model/role/reasoning effort). Set `taskPlanning: captain` when the Captain should derive the DAG from the user's goal; omit it or set `taskPlanning: seed` to keep a fixed template workflow:\n\n```yaml\nprofiles:\n demo-delivery:\n description: Ship a small feature\n protocol: Discuss requirements, review, test, then prepare release; do not deploy automatically.\n members:\n - name: analyst\n model: gpt-5.6-sol\n role: Analyze requirements\n - name: implementer\n model: gpt-5.6-terra\n role: Implement the approved solution\n tasks:\n - id: requirements\n subject: Requirements discussion\n assignee: analyst\n - id: implementation\n subject: Implement solution\n assignee: implementer\n dependencies: [requirements]\n```\n\nUse an explicit profile flag: `/agent-teams --profile demo-delivery implement the feature`. The first ordinary token is never treated as an implicit profile. Normal command runs call `agent_teams_create({ profile, approval: \"required\" })`: the roster and seed/Captain-designed DAG remain staged, no child session is created, and no task is claimed. Edit the plan in the activity panel using the host model catalog, return to chat so the Captain asks what to revise and then atomically updates the same draft, discard it, or click **Approve & Run**. Return/discard actions cancel any planning turn still running; discard also parks model-facing context that forbids silently creating a replacement team. Approval resolves the final provider/model/reasoning choices, commits the roster, and creates each member session only when its first task is ready. A running team is stopped from its own panel header through a confirmation dialog rather than from the composer. Direct tool clients may pass `approval: \"automatic\"` for the legacy immediate path. Failed review/test tasks do not unlock downstream work; automatic repair/review tasks do not depend on the failed review.\n\n## License\n\n[MIT](./LICENSE)\n","readmeFilename":"README.md"}
![\"dsh-agent-teams](\"./assets/readme/hero.svg\")
\n![\"Recommended](\"https://img.shields.io/badge/recommended%20by-dshfind-FFD700?style=flat-square\"){kind=icon}
![\"npm](\"https://img.shields.io/npm/v/@nanmicoder/dsh-agent-teams?style=flat-square&color=5B4CF0\")
![\"MIT](\"https://img.shields.io/badge/license-MIT-0B7285?style=flat-square\"){kind=icon}
![\"DSH](\"https://img.shields.io/badge/DSH-Web%20%2B%20Headless-5B4CF0?style=flat-square\"){kind=icon}
![\"AgentTeams](\"./assets/readme/workspace.png\")
\n