Docs MCP Server

What is Docs MCP Server?

Docs MCP Server is a Model Context Protocol (MCP) server that searches the latest documentation for a given query and library. It supports searching docs for langchain, openai, and llama-index.

How to use Docs MCP Server?

Install Python 3.10+, MCP SDK 1.2.0+, and the uv package manager. Run uv run main.py to start the server, then configure Claude Desktop via its claude_desktop_config.json to connect.

Key features of Docs MCP Server

• Searches latest docs for a given query and library
• Supports langchain, openai, and llama-index
• Lightweight MCP server exposing tools via standardized protocol
• Uses uv for project management and dependency handling
• Requires Python 3.10 or higher

Use cases of Docs MCP Server

• Retrieving up-to-date API documentation for langchain inside an LLM tool
• Searching OpenAI documentation for specific function signatures
• Looking up llama-index code examples through a natural-language query

FAQ from Docs MCP Server

What does the tool function take as arguments?

It takes a query (e.g. "Chroma DB") and a library (e.g. "langchain"). It returns text from the docs.

What are the system requirements?

Python 3.10 or higher, MCP SDK 1.2.0 or higher, and the uv package manager.

How do I connect the server to Claude Desktop?

Edit ~/Library/Application Support/Claude/claude_desktop_config.json to add the server command with the absolute path to your project, then restart Claude Desktop.

Does the server support authentication or remote transport?

The README only describes a local Unix-based setup with Claude Desktop; no authentication or remote transport details are provided.

What libraries can I search docs for?

The server supports langchain, openai, and llama-index.