{
  "markdown": "# agentdocs\n\n**Fresh, source-verified docs for building MCP servers and Claude agents — served straight to your AI coding agent over MCP.**\n\n[![npm](https://img.shields.io/npm/v/mcp-agentdocs)](https://www.npmjs.com/package/mcp-agentdocs)\n[![license: MIT](https://img.shields.io/badge/license-MIT-green)](./LICENSE)\n[![Model Context Protocol](https://img.shields.io/badge/MCP-server-blue)](https://modelcontextprotocol.io)\n\nCoding agents hallucinate on fast-moving APIs like the Model Context Protocol and the Claude Agent SDK: training data lags the spec, so you get plausible-but-wrong imports, deprecated patterns, and invented options. `agentdocs` is a local MCP server that gives your agent a curated, **source-verified** corpus to search instead of guessing.\n\n- 🔎 Three tools: `list_topics`, `search_docs`, `get_doc`\n- 📎 Every answer is attributed to its upstream source URL\n- 🏠 Runs **locally** over stdio — no account, no hosting, no data leaves your machine\n- ✅ Each doc carries a `verified` flag set only after human review against the source\n\n## Install & connect\n\nRequires Node.js 18+.\n\n### Claude Code\n\n```bash\nclaude mcp add agentdocs -- npx -y mcp-agentdocs@latest\n```\n\n### Cursor / other MCP clients (`mcp.json`)\n\n```json\n{\n  \"mcpServers\": {\n    \"agentdocs\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"mcp-agentdocs@latest\"]\n    }\n  }\n}\n```\n\nThen ask your agent things like *\"check agentdocs for how MCP stdio handles logging\"* or\n*\"search agentdocs for the Streamable HTTP session header\"* — it will call the tools and\nanswer from the verified corpus.\n\n## Tools\n\n| Tool | Use it to |\n|------|-----------|\n| `list_topics` | See every curated doc, its topic, source URL, and verified status |\n| `search_docs(query, topic?)` | Get ranked, source-attributed snippets for a question |\n| `get_doc(id)` | Read the full verified markdown of one document |\n\n## What's covered (v1)\n\nThe Model Context Protocol (spec 2025-11-25) and building MCP servers / Claude agents.\nCorpus grows over time; run `list_topics` for the current set.\n\n## Freshness\n\nEvery document records the date it was last verified against its primary source. Sources\nare diffed on a regular cadence; when one drifts, the affected doc is re-verified and a new\nversion is published. Using `mcp-agentdocs@latest` (as above) keeps you on the current corpus.\nSee [CHANGELOG.md](./CHANGELOG.md) for what changed and when.\n\n## Development\n\n```bash\nnpm install\nnpm run build      # tsc -> dist/\nnpm test           # vitest: retrieval, manifest integrity, unicode safety\nnpm run ingest     # refresh sources into corpus/_staging/ for human review\n```\n\nCuration is the point: `npm run ingest` fetches upstream sources for review but never\nsilently overwrites a curated doc — a human diffs, promotes, and sets `verified: true`\nin `corpus/manifest.json` only after checking the content against its source. Run\n`npm run check-freshness` to flag docs that have drifted or are due for re-verification.\n\n## License\n\nMIT\n",
  "bytes": 2989,
  "sha": "8510ae88afbabafae46b1527d680b3b7358c6896e1e3c2106d2f36ac79c1b65d",
  "repo_slug": "titov-d/mcp-agentdocs",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_titov_d_agentdocs_11c3efd6/readme"
}