{"_id":"@agent-smith/feat-agents","_rev":"5-87f90f0645c1fe58b77ec14f01c9e394","name":"@agent-smith/feat-agents","dist-tags":{"latest":"0.0.5"},"versions":{"0.0.1":{"name":"@agent-smith/feat-agents","version":"0.0.1","license":"MIT","_id":"@agent-smith/feat-agents@0.0.1","maintainers":[{"name":"synw","email":"synwx@protonmail.com"}],"homepage":"https://github.com/lynxai-team/agent-smith-plugins#readme","bugs":{"url":"https://github.com/lynxai-team/agent-smith-plugins/issues"},"dist":{"shasum":"eab3cd549c6fc40f1bda82665138d7430b951326","tarball":"https://registry.npmjs.org/@agent-smith/feat-agents/-/feat-agents-0.0.1.tgz","fileCount":39,"integrity":"sha512-ssMzhkTgoTykd9T4NQFlLSWQNQq636Kb16oaeMPp21I/DLeUby7WIU2yHwBefds5oc+4sJkie0CZJvbHrpXS2A==","signatures":[{"sig":"MEUCIAtYdOH15oPc7cBFDB2DgYgIhNCc251PDuIy0nQeBuztAiEAmmOFKFBJphuAQpMnmQaTwQDnDP6mjigWgLjZ5bThEvE=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":76948},"type":"module","gitHead":"db12e263831982b12142d1374f58f446423fd7fa","_npmUser":{"name":"synw","email":"synwx@protonmail.com"},"repository":{"url":"git+https://github.com/lynxai-team/agent-smith-plugins.git","type":"git"},"_npmVersion":"11.13.0","description":"AI agents for Agent Smith","directories":{},"_nodeVersion":"24.11.0","publishConfig":{"access":"public","registry":"https://registry.npmjs.org/"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/feat-agents_0.0.1_1781784006860_0.4045449845459459","host":"s3://npm-registry-packages-npm-production"}},"0.0.2":{"name":"@agent-smith/feat-agents","version":"0.0.2","license":"MIT","_id":"@agent-smith/feat-agents@0.0.2","maintainers":[{"name":"synw","email":"synwx@protonmail.com"}],"homepage":"https://github.com/lynxai-team/agent-smith-plugins#readme","bugs":{"url":"https://github.com/lynxai-team/agent-smith-plugins/issues"},"dist":{"shasum":"d51e71cfcb756aef1d5ce00b2a681e987aa28616","tarball":"https://registry.npmjs.org/@agent-smith/feat-agents/-/feat-agents-0.0.2.tgz","fileCount":43,"integrity":"sha512-PB11lu/3bpFmVAlGaSOAtZ7FSF0QPK+S0xYVxXabMUvfK1YvHtdxtr33tQnm33+2wOG9H6pW5OOUMLNarxD3Nw==","signatures":[{"sig":"MEUCIHF4+zrie6OHxfJdlPECu/nSWPbDYkE/MzMcw8W/gnW5AiEA6LzUN19q2G/QvAZTPq/F4FrKbwRrWZFWwQOeiifDefc=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":102811},"type":"module","gitHead":"c123eb57afa103b3dfe6222ea65ecb30e16aa5c3","_npmUser":{"name":"synw","email":"synwx@protonmail.com"},"repository":{"url":"git+https://github.com/lynxai-team/agent-smith-plugins.git","type":"git"},"_npmVersion":"11.13.0","description":"AI agents for Agent Smith","directories":{},"_nodeVersion":"24.11.0","publishConfig":{"access":"public","registry":"https://registry.npmjs.org/"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/feat-agents_0.0.2_1782036125756_0.4407892675548539","host":"s3://npm-registry-packages-npm-production"}},"0.0.3":{"name":"@agent-smith/feat-agents","version":"0.0.3","license":"MIT","_id":"@agent-smith/feat-agents@0.0.3","maintainers":[{"name":"synw","email":"synwx@protonmail.com"}],"homepage":"https://github.com/lynxai-team/agent-smith-plugins#readme","bugs":{"url":"https://github.com/lynxai-team/agent-smith-plugins/issues"},"dist":{"shasum":"79c77b1cb371c5f9e263a667e0507c81bfb1e9d8","tarball":"https://registry.npmjs.org/@agent-smith/feat-agents/-/feat-agents-0.0.3.tgz","fileCount":43,"integrity":"sha512-Vqlep8UPw0lM8v3o9Gzz/7O5nbMuZFsJcIDF+yVaDsMOs0SSA0oOGqm/zR7HmUau4kQpabzRryIXVckEvtnpWw==","signatures":[{"sig":"MEYCIQCn5S3pPTv3ipGD1Bq8lDpxp5Wn1XV5joacY0peT85ALQIhANrNqFGHgWPUYuyb+dpYmRE8ga02Y3aKcowFa2ros2Oy","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":103151},"type":"module","gitHead":"0749eb0dba48cd5201b074a3a2e60e7bb962d6eb","_npmUser":{"name":"synw","email":"synwx@protonmail.com"},"repository":{"url":"git+https://github.com/lynxai-team/agent-smith-plugins.git","type":"git"},"_npmVersion":"11.18.0","description":"AI agents for Agent Smith","directories":{},"_nodeVersion":"24.11.0","publishConfig":{"access":"public","registry":"https://registry.npmjs.org/"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/feat-agents_0.0.3_1785839308265_0.5805138072192417","host":"s3://npm-registry-packages-npm-production"}},"0.0.4":{"name":"@agent-smith/feat-agents","version":"0.0.4","license":"MIT","_id":"@agent-smith/feat-agents@0.0.4","maintainers":[{"name":"synw","email":"synwx@protonmail.com"}],"homepage":"https://github.com/lynxai-team/agent-smith-plugins#readme","bugs":{"url":"https://github.com/lynxai-team/agent-smith-plugins/issues"},"dist":{"shasum":"63d96d84855b7281b986dd004c4380e025b47b4e","tarball":"https://registry.npmjs.org/@agent-smith/feat-agents/-/feat-agents-0.0.4.tgz","fileCount":44,"integrity":"sha512-ybrSrV93VekLhnpDVk2dnvB+4YHSu5THE/H++1ftOaLu6jxOz1AJ6z+XTPyRlgQy3KLnX1vkQvNoTQ7lDUPQlQ==","signatures":[{"sig":"MEUCIQC9faizCOi1gf3WTkTgticBlYJMZiwM6ZZt6FqrijFqYAIgQWL5gzfi8qe7/3l7TeEvyVs1ifOnKxBUJoNd7xlCTvE=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":104198},"type":"module","gitHead":"47151f444f4b2375526d0449ab28d6c432bf0de9","_npmUser":{"name":"synw","email":"synwx@protonmail.com"},"repository":{"url":"git+https://github.com/lynxai-team/agent-smith-plugins.git","type":"git"},"_npmVersion":"11.18.0","description":"AI agents for Agent Smith","directories":{},"_nodeVersion":"24.11.0","publishConfig":{"access":"public","registry":"https://registry.npmjs.org/"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/feat-agents_0.0.4_1786715751638_0.12488703358371112","host":"s3://npm-registry-packages-npm-production"}},"0.0.5":{"name":"@agent-smith/feat-agents","version":"0.0.5","description":"AI agents for Agent Smith","repository":{"type":"git","url":"git+https://github.com/lynxai-team/agent-smith-plugins.git"},"type":"module","publishConfig":{"access":"public","registry":"https://registry.npmjs.org/"},"license":"MIT","gitHead":"5ca6a9cc9d3fcce2025b74ca3e5aee1dee0dda6c","_id":"@agent-smith/feat-agents@0.0.5","bugs":{"url":"https://github.com/lynxai-team/agent-smith-plugins/issues"},"homepage":"https://github.com/lynxai-team/agent-smith-plugins#readme","_nodeVersion":"24.11.0","_npmVersion":"11.18.0","dist":{"integrity":"sha512-p/HcQBimm8tM6an6bn7tNNB03YRss6Z6uco6S0/smF62D19GC4Af6PgBX5Wrl7Wt7NL/SQ4JTyrhZqP6h0TvaA==","shasum":"1d89deb1855e117eb70baabcc41bb056f1ad7c95","tarball":"https://registry.npmjs.org/@agent-smith/feat-agents/-/feat-agents-0.0.5.tgz","fileCount":44,"unpackedSize":104193,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEQCIH9idKd0uluYG1a/8jLHRU3A93kD9zFbuFqzNnMJ6EBzAiAW0gq7zq08GMQoWTYxIbOnmMXuS8Ay9gUoALPqUH0y0g=="}]},"_npmUser":{"name":"synw","email":"synwx@protonmail.com"},"directories":{},"maintainers":[{"name":"synw","email":"synwx@protonmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/feat-agents_0.0.5_1787313446363_0.04054879025874514"},"_hasShrinkwrap":false}},"time":{"created":"2026-06-18T12:00:06.755Z","modified":"2026-08-21T11:57:26.679Z","0.0.1":"2026-06-18T12:00:07.045Z","0.0.2":"2026-06-21T10:02:05.890Z","0.0.3":"2026-08-04T10:28:28.406Z","0.0.4":"2026-08-14T13:55:51.779Z","0.0.5":"2026-08-21T11:57:26.535Z"},"bugs":{"url":"https://github.com/lynxai-team/agent-smith-plugins/issues"},"license":"MIT","homepage":"https://github.com/lynxai-team/agent-smith-plugins#readme","repository":{"type":"git","url":"git+https://github.com/lynxai-team/agent-smith-plugins.git"},"description":"AI agents for Agent Smith","maintainers":[{"name":"synw","email":"synwx@protonmail.com"}],"readme":"# @agent-smith/feat-agents\n\n[![npm](https://img.shields.io/npm/v/@agent-smith/feat-agents)](https://www.npmjs.com/package/@agent-smith/feat-agents)\n\n**Multi-agent coordination system for the Agent Smith toolkit** ā€” 16 specialized AI agents, 4 workflows, and 15+ reusable skills for task management and project automation.\n\nThis package is part of the [Agent Smith](https://github.com/lynxai-team/agent-smith) project, a TypeScript toolkit for running and managing AI agents.\n\n---\n\n## šŸ“– Documentation\n\n### For AI Agents\n- [Codebase Summary](../.agents/documentation/codebase-summary.md) ā€” Architecture, agent definitions, skills, and patterns for the agents plugin\n- [Project Navigation](../../.agents/documentation/project-nav.md) ā€” Map of all Agent Smith repositories and plugins\n\n### For Humans\n- [Plugin Overview](https://lynxai-team.github.io/agent-smith/plugins/) ā€” Complete plugin documentation\n- [GitHub Repository](https://github.com/lynxai-team/agent-smith-plugins) ā€” Source code and contributions\n\n---\n\n## āœØ Features\n\n- šŸ¤– **Coordinator Agent** ā€” Central `agent-smith` agent that decomposes tasks, plans workflows, and delegates to specialized agents via tool calls\n- šŸ” **16 Specialized Agents** ā€” Pre-configured agents for search, code, documentation, SQL queries, assistance, help, state tracking, project management, and collaboration\n- āš” **4 Workflows** ā€” YAML-defined pipelines for config info retrieval, database queries, Q&A sessions, and vision tasks\n- šŸ“‹ **15+ Reusable Skills** ā€” Task creation/execution (solo/team), documentation updates, codebase summaries, and project management\n- šŸ§© **Context Fragments** ā€” Workspace context helpers that provide agents with project structure awareness\n- šŸ”— **Agent Delegation** ā€” Coordinator calls specialized agents sequentially via the `run-agent` tool\n- šŸŽÆ **Model Flexibility** ā€” Supports all compatible LLM models (qwen35b, qwen4b, etc.) with configurable inference parameters\n- šŸ›”ļø **Security Patterns** ā€” Agents operate within a specified workspace directory with controlled tool access\n\n---\n\n## šŸ“¦ Installation\n\n```bash\n# Install globally via npm\nnpm i -g @agent-smith/feat-agents\n```\n\n### Configuration\n\nAdd the plugin to your `config.yml`:\n\n```yaml\nplugins:\n - \"@agent-smith/feat-agents\"\n```\n\nThen update your client configuration:\n\n```bash\nlm conf\n```\n\n---\n\n## šŸš€ Quick Start\n\nLaunch the main coordinator agent:\n\n```bash\nlm agent-smith \"Create documentation for my project\"\n```\n\nThe coordinator will automatically decompose your request and delegate to specialized agents as needed.\n\nFor direct inference (bypassing coordination):\n\n```bash\nlm infer \"Explain how multi-agent coordination works\"\n```\n\n---\n\n## šŸ“‹ Agents Reference\n\n### Core Agents\n\n| Agent | Description |\n|-------|-------------|\n| `agent-smith` | **Main coordinator** ā€” Orchestrates team of agents, decomposes tasks, delegates work |\n| `agent-smith-assistant` | Default assistant agent ā€” Fork of coordinator for general task execution |\n| `agent-smith-light` | Lightweight agent for simpler tasks with reduced context |\n| `infer` | Raw inference agent for direct LLM queries without orchestration |\n\n### Specialized Agents\n\n| Agent | Description |\n|-------|-------------|\n| `agent-smith-search` | Web search and webpage reading ā€” uses ddsearch, read-webpage tools |\n| `agent-smith-code` | Code management agent for programming tasks |\n| `agent-smith-doc` | Documentation management agent for project docs |\n| `agent-smith-sql` | Database query agent ā€” generates SQL from natural language questions |\n| `agent-smith-help` | Help agent with full capabilities |\n| `agent-smith-help-light` | Lightweight help agent |\n| `agent-smith-help-assistant` | Delegated work assistant for help scenarios |\n| `agent-smith-state` | Project state management ā€” checks context and documentation completeness |\n| `agent-smith-project` | Project management agent |\n| `agent-smith-colab` | Collaborative agent (experimental) |\n| `agent-smith-doc-colab` | Documentation collaboration agent |\n| `collaborator` | Worker collaborator agent for role-based tasks |\n\n---\n\n## šŸ”„ Workflows\n\nThe plugin includes 4 workflow definitions for multi-step operations:\n\n| Workflow | Purpose |\n|----------|---------|\n| `agent-smith-config-info` | Configuration information retrieval workflows |\n| `agent-smith-db` | Database query orchestration with SQL agent |\n| `q` | Q&A workflow ā€” adapts pre-query then runs inference agent |\n| `vision` | Vision tasks workflow for image/video analysis |\n\nExecute a workflow:\n\n```bash\nlm workflow agent-smith-config-info\n```\n\n---\n\n## šŸ› ļø Skills Reference\n\nThe plugin provides 15+ reusable skills for AI coding agents, loaded via the `load-skill` tool.\n\n### Task Creation\n| Skill | Description |\n|-------|-------------|\n| `create-task` | Entry point ā€” analyzes request, chooses execution mode, creates task |\n| `create-task-solo` | Single-agent task creation |\n| `create-task-team` | Multi-agent team task creation |\n| `create-task-from-template` | Create tasks from predefined templates |\n| `use-tasks` | Task usage and management skill |\n\n### Task Execution\n| Skill | Description |\n|-------|-------------|\n| `execute-task` | Task execution (uses phase-specific skills) |\n| `execute-task-solo` | Single-agent task execution |\n| `execute-task-team` | Multi-agent team task execution |\n| `execute-task-phase` | Phase-specific execution logic |\n| `task-mode` | Decision skill ā€” evaluates task complexity to choose solo vs team mode |\n\n### Documentation\n| Skill | Description |\n|-------|-------------|\n| `create-readme` | Create comprehensive README files for packages/plugins/repositories |\n| `document-package` | Create or update package documentation in the docsite |\n| `create-project-docs` | Generate project documentation structure |\n| `update-codebase-summary` | Create or update codebase summary documentation |\n| `update-project-nav` | Update the project navigation map |\n\n### Exploration\n| Skill | Description |\n|-------|-------------|\n| `smart-explore` | Navigate and explore module directory trees efficiently |\n\n---\n\n## šŸ“‹ Tasks Reference\n\nThe plugin includes pre-defined tasks ā€” structured work units with goals, plans, and progress tracking for complex multi-step operations. Use the `load-task` tool to load a task into your workspace, then execute it with the appropriate skill.\n\n| Task | Description | Mode | Phases |\n|------|-------------|------|--------|\n| `update-project-doc` | Create or update AI agent navigation documentation for a project (AGENTS.md, decision-tree.md, codebase-summary.md, project-nav.md, etc.) | Solo | 8 phases: Explore ā†’ AGENTS.md ā†’ Per-Repo AGENTS.md ā†’ decision-tree ā†’ project-overview ā†’ codebase-summary ā†’ project-nav ā†’ Cross-Reference |\n\n### Using Tasks\n\nPrompt to execute a task:\n\n```\nexecute the update-project-doc\n\n%execute-task-solo\n```\n\nThe `%execute-task-solo` part loads this skill directly in the prompt, so that the agent has all the execution instructions.\nUse `%execute-task-team` for multi agents tasks. Note: the agent must have the `load-task` tool. When it uses it the task directory is copied to `.agents/tasks/[task-name]` and it's execution and state will be managed from there.\n\nThe agent will execute phases sequentially, verify success criteria after each step, and update `state.md` to track progress.\n\n### Task Structure\n\nEach task contains:\n- `goals.md` ā€” Goals and success criteria\n- `plan.md` ā€” Step-by-step execution plan with verifiable success criteria\n- `state.md` ā€” Progress tracking (updated after each phase)\n- `notes.md` ā€” Optional context, rules, and related skills\n\n---\n\n## šŸ“ Structure\n\n```\nagents/\nā”œā”€ā”€ package.json # Plugin metadata (@agent-smith/feat-agents)\nā””ā”€ā”€ dist/\n ā”œā”€ā”€ agents/ # 16 agent YAML definitions\n ā”‚ ā”œā”€ā”€ agent-smith.yml # Main coordinator agent\n ā”‚ ā”œā”€ā”€ agent-smith-assistant.yml\n ā”‚ ā”œā”€ā”€ agent-smith-search.yml\n ā”‚ ā”œā”€ā”€ agent-smith-code.yml\n ā”‚ ā”œā”€ā”€ agent-smith-doc.yml\n ā”‚ ā”œā”€ā”€ agent-smith-sql.yml\n ā”‚ ā”œā”€ā”€ agent-smith-help.yml\n ā”‚ ā””ā”€ā”€ ... (16 total)\n ā”œā”€ā”€ workflows/ # 4 workflow definitions\n ā”‚ ā”œā”€ā”€ agent-smith-config-info.yml\n ā”‚ ā”œā”€ā”€ agent-smith-db.yml\n ā”‚ ā”œā”€ā”€ q.yml\n ā”‚ ā””ā”€ā”€ vision.yml\n ā”œā”€ā”€ skills/ # 15+ skill modules\n ā”‚ ā”œā”€ā”€ create-readme/SKILL.md\n ā”‚ ā”œā”€ā”€ create-task/SKILL.md\n ā”‚ ā”œā”€ā”€ smart-explore/SKILL.md\n ā”‚ ā””ā”€ā”€ ... (16 total)\n ā”œā”€ā”€ tasks/ # Pre-defined task definitions\n ā”‚ ā””ā”€ā”€ update-project-doc/ # Project documentation task\n ā”‚ ā”œā”€ā”€ goals.md\n ā”‚ ā”œā”€ā”€ plan.md\n ā”‚ ā”œā”€ā”€ state.md\n ā”‚ ā””ā”€ā”€ notes.md\n ā””ā”€ā”€ fragments/ # Context helper files\n ā”œā”€ā”€ workspace.md # Workspace context info\n ā””ā”€ā”€ ctx-helper-files.md # Context file references\n```\n\n---\n\n## šŸ”§ Architecture\n\n### Agent Coordination Pattern\n\nThe `agent-smith` coordinator agent follows a structured workflow:\n\n1. **Analyze** the user request\n2. **Decompose** into subtasks if necessary\n3. **Plan** the work sequence\n4. **Delegate** to specialized agents via `run-agent` tool (one at a time)\n5. **Synthesize** results from agent responses\n\n### Available Tools\n\nAgents have access to these tools:\n\n| Tool | Description |\n|------|-------------|\n| `readfile` | Read file contents |\n| `edit-search-replace` | Search and replace in files |\n| `shell` | Execute shell commands |\n| `python` | Execute Python code |\n| `run-agent` | Call other agents for delegated work |\n| `load-skill` | Load skill modules for specific tasks |\n| `load-task` | Load task definitions for complex operations |\n| `notify-user` | Send notifications to the user |\n\n### Context Fragments\n\nAgents can include project context using file references in YAML:\n\n```yaml\ntemplate:\n system: |-\n {file:../fragments/workspace.md}\n {file:../fragments/ctx-helper-files.md}\n```\n\n### Sample Agent Definition\n\nEach agent is defined as a YAML file:\n\n```yaml\nname: infer\ndescription: Run a raw inference query\ntype: \"agent-smith\"\nprompt: |-\n {prompt}\nmodel: qwen4b\ninferParams:\n top_p: 0.95\n top_k: 20\n min_p: 0\n temperature: 0.6\n repeat_penalty: 1\n```\n\n---\n\n## āš™ļø Configuration\n\n### Model Parameters\n\nAgents support configurable inference parameters:\n\n```yaml\ninferParams:\n min_p: 0\n top_k: 20\n top_p: 0.85\n temperature: 0.6\n repetition_penalty: 1\n presence_penalty: 1.5\n chat_template_kwargs:\n enable_thinking: true\n preserve_thinking: true\n```\n\n### Variables\n\nAgents can require runtime variables:\n\n```yaml\nvariables:\n required:\n workspace:\n description: The local directory path where to operate\n```\n\n### Customizing Agent Behavior\n\nTo customize an agent, copy the YAML from `dist/agents/` to your local features directory and modify it. The coordinator uses a system prompt that lists available agents and their purposes:\n\n```yaml\ntemplate:\n system: |-\n You are Agent Smith, an AI coordinator agent. Your job is to organise\n the workflow of other agents. Available agents:\n - `agent-smith-assistant`: default assistant\n - `agent-smith-search`: web search and reading\n {file:../fragments/workspace.md}\n```\n\n---\n\n## šŸ”— Related Packages\n\n| Package | Description |\n|---------|-------------|\n| [`@agent-smith/core`](https://github.com/lynxai-team/agent-smith) | Core framework ā€” workflow engine, agent runtime, tool integration |\n| [`@agent-smith/cli`](https://github.com/lynxai-team/agent-smith) | Terminal client that consumes plugins |\n| [`@agent-smith/feat-git`](https://www.npmjs.com/package/@agent-smith/feat-git) | Git operations with AI-powered commit messages |\n| [`@agent-smith/feat-sqlite`](https://www.npmjs.com/package/@agent-smith/feat-sqlite) | SQLite database operations |\n| [`@agent-smith/feat-search`](https://www.npmjs.com/package/@agent-smith/feat-search) | Web search and crawling tools (used by search agent) |\n| [`@agent-smith/feat-shell`](https://www.npmjs.com/package/@agent-smith/feat-shell) | Sandboxed shell/Python execution in Docker containers |\n| [`@agent-smith/feat-fs`](https://www.npmjs.com/package/@agent-smith/feat-fs) | Filesystem operations with path authorization |\n\n---\n\n## āš ļø Important Notes\n\n- **No runtime code** ā€” Agents, workflows, and skills are YAML/Markdown files loaded by the `@agent-smith/core` framework at runtime\n- **YAML-based configuration** ā€” All agent definitions use YAML for easy customization\n- **Sequential delegation** ā€” The coordinator calls one agent at a time for task completion\n- **Model flexibility** ā€” Supports all compatible LLM models with configurable inference parameters\n- **Workspace context** ā€” Agents operate within a specified workspace directory (required variable)\n- **Requires core framework** ā€” This plugin depends on `@agent-smith/core` for execution\n\n---\n\n## šŸ“„ License\n\nMIT\n","readmeFilename":"README.md"}