# langchain-ai/mcpdoc

Expose llms-txt to IDEs for development

Repository: https://github.com/langchain-ai/mcpdoc
Canonical: https://ross.abutalabs.com/products/mcpdoc
Language: Python
License: MIT
License Family: permissive
Topics: agents, claude-code, cursor, ide, llms, llms-txt
Last push: 2026-08-20T06:34:29+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 40, longevity 38
- inputs: {"age_days": 533, "days_push": 13, "days_rel": 407, "gap_med": 2, "n_releases_24m": 10}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1031, forks 128 (observed 2026-08-28T04:03:18.128284+00:00)

## What it is
An MCP server that exposes llms.txt documentation indexes to IDEs and AI coding apps like Cursor, Windsurf, and Claude Code. It provides a fetch_docs tool with domain access controls so users can audit tool calls and the documentation context returned to their LLM.

## Use cases
- give cursor access to langchain docs via llms.txt
- serve llms.txt documentation to claude code through mcp
- audit documentation fetches made by ai coding assistants
- restrict which domains an ai ide can fetch docs from
- add library documentation context to windsurf
- make project docs available to llm agents in the ide

## When to choose
- you use Cursor, Windsurf, or Claude Code and want controlled, auditable access to llms.txt docs
- you need to restrict documentation fetching to specific domains for security
- you want a user-defined list of documentation sources for AI-assisted coding

## When to avoid
- you need general web search or scraping rather than llms.txt doc retrieval
- your IDE already handles llms.txt retrieval and you don't need auditing or domain controls

## Facets
- artifact type: service
- maturity: active
- function: mcp, documentation, search-engine, developer-tools
- domain: developer-tools, documentation, large-language-models
- platform: python, cli, self-hosted
- tags: llms-txt, mcp-server, ide-integration, cursor, claude-code, windsurf, docs-retrieval, ai-agents

## Member repositories
- langchain-ai/mcpdoc (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:18.128284+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:07:27.712057+00:00, confidence not recorded.
  - readme: https://github.com/langchain-ai/mcpdoc (fetched 2026-08-28T04:03:18.128284+00:00, sha 758265258116)
- Data as of 2026-08-30T08:39:29.467469+00:00.