MCP-Server-for-Documentation

MCP-Server-for-Documentation

0
research_and_datacloud_platformsdeveloper_tools

The Docs MCP Server provides an implementation of the Model Context Protocol (MCP), allowing standardized context provision to LLMs. It supports multiple libraries and follows a client-server architecture to access both local and remote data sources.

Docs MCP Server

This repository contains an implementation of a Model Context Protocol (MCP) server.

Features

  • Searches the latest documentation for a given query and library.
  • Supports langchain, openai, and llama-index.
  • Follows a client-server architecture allowing host applications to connect to multiple servers.
  • Provides three main capabilities: Resources, Tools, and Prompts.

System Requirements

  • Python 3.10 or higher
  • MCP SDK 1.2.0 or higher
  • uv package manager

Getting Started

  1. Install uv Package Manager.
  2. Setup the project with a virtual environment and install dependencies.
  3. Create the server implementation and run the server.

Connecting to Claude Desktop

  • Install Claude Desktop.
  • Configure to use your MCP server and restart Claude Desktop.

Troubleshooting

  • Ensure configuration file path and permissions are correct.
  • Verify uv installation and accessibility.