MCP Documentation Server

@esakrissa

Repository

Basic information

About MCP Documentation Server

Model Context Protocol documentation server for LangGraph and MCP.

#mcp-doc#documentation #mcp

MCP Documentation Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "mcp-doc": {
      "command": "python",
      "args": [
        "-m",
        "venv",
        "venv"
      ]
    }
  }
}

MCP Documentation Server tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is MCP Documentation Server?

The MCP Documentation Server is a customized version of the MCP documentation server that enables integration between LLM applications (like Cursor, Claude Desktop, Windsurf) and documentation sources via the Model Context Protocol. It provides access to specific documentation files (langgraph.txt and mcp.txt) and tools to fetch documentation from URLs within those files.

How to use MCP Documentation Server?

Clone the repository, create a virtual environment, install the package in development mode (pip install -e .), then run the server using the installed command mcpdoc with a JSON config file and transport options (SSE or stdio). For IDE integration like Cursor, add a configuration to ~/.cursor/mcp.json using UV and set custom instructions to guide the LLM to use the server's tools.

Key features of MCP Documentation Server

Integration with MCP host applications like Cursor, Claude Desktop, Windsurf
Access to LangGraph and MCP documentation files
Tools to fetch documentation from URLs listed in those files
Support for SSE and stdio transport protocols
Configurable allowed domains for security
Easy setup via pip or UV

Use cases of MCP Documentation Server

Answering questions about LangGraph by fetching relevant documentation
Answering questions about Model Context Protocol (MCP) using its docs
Enabling LLM applications to automatically retrieve and reflect on documentation URLs
Customizing allowed domains to restrict or permit documentation sources

Frequently asked questions

What documentation sources are available?

The server is currently set up for LangGraph Documentation and MCP Documentation, sourced from specific URLs.

How do I run the server?

You can run the server using the `mcpdoc` command with a config file, transport (SSE or stdio), port, and host. Alternatively, you can use UV with the `uvx --from mcpdoc mcpdoc` command.

How do I integrate with Cursor?

Add a configuration entry to `~/.cursor/mcp.json` and provide custom instructions in Cursor to call `list_doc_sources` and `fetch_docs` tools.

What security controls are available?

Strict domain access controls: remote documentation files allow only the specific domain by default; local files allow none. Use `--allowed-domains` to explicitly add domains or `*` to allow all.

What is the relationship to the original mcpdoc?

This server is a modified version of the original mcpdoc by LangChain AI, customized to provide focused access to LangGraph and MCP documentation.

Comments

More Memory & Knowledge MCP servers

ctxfile

Local-first MCP server that snapshots your project's working state into one context object. Open-core, privacy-first.

Agimem

Persistent memory for AI agents. Hosted MCP server with capsule-scoped key/value storage, tags, TTL expiration, bulk ops, and RBAC.

Azquo

Azquo Holdings Ltd

Exact, shared memory for AI agents: a hierarchical database where totals always add up, every write is attributed to the agent and person behind it, and any figure traces back to its source file. Agents create databases,

Sooveryn

PANDAPROG

Sooveryn gives Claude Code a team of AI personas that remembers your project. Soov, the partner and team lead, keeps your decisions and how you work in a lasting memory, and finds them again by words and by meaning, from

agentleFS

One permissioned folder your team and your AI agents both log into.