Documentation MCP

🚀 Documentation MCP Server 📚🔍

A Model Context Protocol (MCP) server that enables Claude to directly search and access documentation for popular libraries (such as LangChain, LlamaIndex, and OpenAI) within conversations.

🚀 Quick Start

This MCP server allows Claude to search and access documentation for popular AI libraries directly. It standardizes the way applications provide context to large - language models, acting as a universal connector for AI assistants.

✨ Features

• Documentation Search Tool: Search for documentation in popular AI libraries.
• Supported Libraries:
  • LangChain 🔗
  • LlamaIndex 🦙
  • OpenAI 🤖
• Intelligent Extraction: Intelligently parse HTML content and extract the most relevant information.
• Configurable Results: Limit the length of the returned text according to requirements.

📦 Installation

1. Install the uv Package Manager

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

2. Clone and Set Up the Project

# Clone or download the project
cd documentation

# Create and activate a virtual environment
uv venv
# On Windows:
.venv\Scripts\activate
# On macOS/Linux:
source .venv/bin/activate

# Install dependencies
uv pip install -e .

3. Configure the Serper API Key

Create a .env file in the project directory and add your Serper API key:

SERPER_API_KEY=your_serper_api_key_here

You can register at serper.dev to obtain a Serper API key.

4. Configure Claude Desktop

Edit your Claude desktop configuration file (the location varies by system):

• Windows: /C:/Users/[Your Username]/AppData/Roaming/Claude/claude_desktop_config.json
• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Add the following to the mcpServers section:

"documentation": {
  "command": "uv",
  "args": [
    "--directory",
    "/ABSOLUTE/PATH/TO/YOUR/documentation",
    "run",
    "main.py"
  ]
}

Replace /ABSOLUTE/PATH/TO/YOUR/documentation with the path to your documentation.

5. Start the Server

Run the following command to start the MCP server:

uv run main.py

💻 Usage Examples

Basic Usage

After starting the server, Claude can search and access documentation for specified libraries by calling the get_docs function. For example:

print(get_docs("langchain"))  # Search for LangChain-related documentation

Advanced Usage

The get_docs function can be used with different queries to search for various content.

print(get_docs("llama"))  # Search for LlamaIndex-related content

📚 Documentation

What is MCP? 🤔

MCP (Model Context Protocol) is an open protocol that standardizes the way applications provide context to large - language models. It can be considered a universal connector that enables AI assistants like Claude to directly access external data sources and tools.

How it Works 🛠️

1. The server uses the Serper API to perform Google searches with site - specific queries.
2. Retrieve the content of the search results.
3. BeautifulSoup extracts the most relevant information from the main content area.
4. Claude can access this information through the get_docs tool.

System Requirements 🖥️

• Python 3.11 or higher
• uv package manager
• Serper API key

Tool Parameters 🛠️

• get_docs(query: str): Returns relevant documentation content based on the query.
• Parameters:
  • query (required): The keyword or phrase to search for.

Notes

⚠️ Important Note

Ensure that all dependencies are installed and the Python version meets the requirements. Make sure the configuration file path is correct so that Claude can recognize the MCP server. If you encounter any issues, check the console output or refer to the documentation.