GitMCP

Features • Usage • How It Works • Examples • FAQ • Privacy • Contributing • License

What is GitMCP?

GitMCP is a free, open-source service that seamlessly transforms any GitHub project into a remote Model Context Protocol (MCP) endpoint, enabling AI assistants to access and understand the project's documentation effortlessly.

Features

• Empower AI with GitHub Project Access: Direct your AI assistant to GitMCP for instant access to any GitHub project's documentation, complete with semantic search capabilities to optimize token usage.
• Zero Setup Required: No configurations or modifications needed — GitMCP works out of the box.
• Completely Free and Private: GitMCP is free. We don't collect any personally identifiable information or queries. Plus, you can host it yourself!

Getting Started (Usage)

To make your GitHub repository accessible to AI assistants via GitMCP, use the following URL formats:

• For GitHub repositories: gitmcp.io/{owner}/{repo}
• For GitHub Pages sites: {owner}.gitmcp.io/{repo}
• Dynamic endpoint: gitmcp.io/docs

Congratulations! The chosen GitHub project is now fully accessible to your AI.

Replace {owner} with your GitHub username or organization name and {repo} with your repository name. Once configured, your AI assistant can access the project's documentation seamlessly. The dynamic endpoint doesn't require a pre-defined repository. When used, your AI assistant can dynamically input any GitHub repository to enjoy GitMCP's features.

How It Works

GitMCP serves as a bridge between your GitHub repository's documentation and AI assistants by implementing the Model Context Protocol (MCP). When an AI assistant requires information from your repository, it sends a request to GitMCP. GitMCP retrieves the relevant content and provides semantic search capabilities, ensuring efficient and accurate information delivery.

Examples

Here are some examples of how to use GitMCP with different repositories:

• Example 1: For the repository https://github.com/octocat/Hello-World, use: https://gitmcp.io/octocat/Hello-World
• Example 2: For the GitHub Pages site langchain-ai.github.io/langgraph, use: https://langchain-ai.gitmcp.io/langgraph
• Example 3: Use the generic gitmcp.io/docs endpoint for your AI to dynamically select a repository

These URLs enable AI assistants to access and interact with the project's documentation through GitMCP.

Tools

GitMCP provides a set of tools that can be used to access and interact with the project's documentation.

fetch_<repo-name>_documentation:

Fetches the documentation for the {owner}/{repo} GitHub repository (as extracted from the URL: gitmcp.io/{owner}/{repo} or {owner}.gitmcp.io/{repo}). Useful for general questions. Retrieves the llms.txt file and falls back to README.md or other pages if the former is unavailable.

search_<repo-name>_documentation:

It searches the repository's documentation by providing a query. This is useful for specific questions. It uses semantic search to find the most relevant documentation. This mitigates the cost of a large documentation set that cannot be provided as direct context to LLMs.

fetch_url_content:

Fetches content from a URL that was referenced in previously retrieved documentation and converts it to markdown. This enables the AI assistant to access and process additional resources linked within the project documentation (extremely handy for llms.txt).

search_<repo-name>_code:

Searches code in the repository using the GitHub code search API. Provides results of code matching the search query, helping AI assistants find relevant implementation details and examples within the codebase.

Note: In the case of a generic gitmcp.com/docs usage, the tools are called fetch_generic_documentation, search_generic_code, and search_generic_documentation. They receive additional owner and repo arguments.

FAQ

What is the Model Context Protocol?

The Model Context Protocol is a standard that allows AI assistants to request and receive additional context from external sources in a structured manner, enhancing their understanding and performance.

Does GitMCP work with any AI assistant?

Yes, GitMCP is compatible with any AI assistant supporting the Model Context Protocol, including tools like Cursor, VSCode, Claude, etc.

Is GitMCP compatible with all GitHub projects?

Absolutely! GitMCP works with any public GitHub repository without requiring any modifications. It prioritizes the llms.txt file and falls back to README.md or other pages if the former is unavailable. Future updates aim to support additional documentation methods and even generate content dynamically.

Does GitMCP cost money?

No, GitMCP is a free service to the community with no associated costs.

Privacy

GitMCP is deeply committed to its users' privacy. The service doesn't have access to or store any personally identifiable information as it doesn't require authentication. In addition, it doesn't store any queries sent by the agents. Moreover, as GitMCP is an open-source project, it can be deployed independently in your environment.

GitMCP only accesses content that is already publicly available and only when queried by a user. GitMCP does not automatically scrape repositories. Before accessing any GitHub Pages site, the code checks for robots.txt rules and follows the directives set by site owners, allowing them to opt out. Please note that GitMCP doesn't permanently store data regarding the GitHub projects or their content.

Contributing

We welcome contributions! Please take a look at our contribution guidelines.

License

This project is licensed under the MIT License.

Disclaimer

GitMCP is provided "as is" without warranty of any kind. While we strive to ensure the reliability and security of our service, we are not responsible for any damages or issues that may arise from its use. GitHub projects accessed through GitMCP are subject to their respective owners' terms and conditions. GitMCP is not affiliated with GitHub or any of the mentioned AI tools.