MCP Server For Documentation

🚀 Document MCP Server

This repository implements a Model Context Protocol (MCP) server.

The MCP server """ Searches for the latest documents based on the given query and library. It supports langchain, openai, and llama-index.

Parameters: query: The content to search for (e.g., "Chroma DB") library: The library to search in (e.g., "langchain")

Returns: The text from the documents """

🔍 What is MCP?

MCP (Model Context Protocol) is an open protocol that standardizes how applications provide context to LLMs. Imagine MCP as a USB-C interface, offering a standard way for AI applications to connect different data sources and tools.

✨ Key Advantages

• A growing list of pre-built integrations that your LLM can directly connect to
• Flexibility when switching LLM providers and vendors
• Best practices for protecting data in the infrastructure

🏗️ Architecture Overview

MCP adopts a client-server architecture, where host applications can connect to multiple servers:

• MCP Host: Such as Claude Desktop, IDEs, or AI tools, these programs want to access data via MCP.
• MCP Client: A protocol client that maintains a 1:1 connection with the server.
• MCP Server: A lightweight program that exposes specific functions through the standardized Model Context Protocol.
• Data Sources: Local (files, databases) and remote services (APIs), etc., which the MCP server can access.

📚 Core MCP Concepts

MCP servers can provide three main types of capabilities:

• Resources: Data similar to files that clients can read (e.g., API responses or file contents).
• Tools: Functions that LLMs can call (requiring user approval).
• Prompts: Pre-written templates to assist users in completing specific tasks.

🛠️ System Requirements

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

🚀 Quick Start

📦 Install the uv Package Manager

On MacOS/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Please make sure to restart your terminal afterward to pick up the uv command.

📁 Project Setup

1. Create and initialize the project:

# Create a new directory for our project
uv init mcp-server
cd mcp-server

# Create a virtual environment and activate it
uv venv
source .venv/bin/activate  # On Windows, use: .venv\Scripts\activate

# Install dependencies
uv add "mcp[cli]" httpx

2. Create the server implementation file:

touch main.py

▶️ Run the Server

1. Start the MCP server:

uv run main.py

2. The server will start and be ready to accept connections.

🔗 Connect to Claude Desktop

1. Install Claude Desktop from the official website.
2. Configure Claude Desktop to use your MCP server:

Edit ~/Library/Application Support/Claude/claude_desktop_config.json:

{
    "mcpServers": {
        "mcp-server": {
            "command": "uv",  # It's better to use the absolute path of the uv command
            "args": [
                "--directory",
                "/ABSOLUTE/PATH/TO/YOUR/mcp-server",
                "run",
                "main.py"
            ]
        }
    }
}

3. Restart Claude Desktop.

🐞 Troubleshooting

If your server is not detected by Claude Desktop:

1. Check the configuration file path and permissions.
2. Ensure the absolute paths in the configuration are correct.
3. Make sure uv is installed correctly and accessible.
4. Check for any error messages in the Claude Desktop logs.