MyMLH

You can use this MCP server to securely access the MyMLH API on behalf of users, enabling AI assistants and MCP clients to fetch user profiles, education, employment history, and more with OAuth-backed authentication and automatic token management. It runs at the edge on Cloudflare for low latency and straightforward deployment.

How to use

Connect your MCP client to the provided HTTP MCP endpoint or use the local stdio option if your environment cannot use the HTTP transport. Once connected and authenticated, you can fetch the current user’s MyMLH profile, retrieve profile details, and perform actions that rely on MyMLH data. Ensure your client follows the OAuth flow so tokens refresh automatically and securely.

How to install

Prerequisites: Node.js and npm or npx installed on your system. You should also have a compatible MCP client that supports Streamable HTTP with OAuth.

Step 1: Decide how you want to run the server. You can use the remote HTTP MCP endpoint or run a local stdio wrapper depending on your environment.

HTTP MCP endpoint configuration

{
  "mcpServers": {
    "mymlh": {
      "type": "http",
      "url": "https://mymlh-mcp.git.ci/mcp",
      "args": []
    }
  }
}

Fallback stdio configuration

{
  "mcpServers": {
    "mymlh": {
      "type": "stdio",
      "name": "mymlh",
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mymlh-mcp.git.ci/mcp"
      ],
      "env": []
    }
  }
}

Additional setup notes

If your client does not support Streamable HTTP with OAuth, prefer the HTTP endpoint when possible. For environments that require a local wrapper, the stdio configuration forwards requests over HTTP behind the scenes to ensure compatibility.

Security and testing

The server uses OAuth to authenticate users with the MyMLH API. Token refreshes are handled automatically and tokens are stored securely by the server. To test the integration, connect with an MCP inspector or the Cloudflare AI Playground and follow the authentication flow.

Example usage

Ask your AI assistant to fetch user data or generate documents using MyMLH data, for example: “Get my MyMLH user info.” or “Generate a resume using my MyMLH profile.”

Deploying your own instance

You can deploy a personal MCP server instance to Cloudflare for full control and private usage. Follow the steps in the deployment guide to publish to your Cloudflare account.

Contributing

Contributions are welcome. To contribute, follow the project’s contributing guidelines and set up your development environment as described in the CONTRIBUTING guide.

License

MIT License

Available tools

mymlh_get_user

Fetch the current MyMLH user profile.

mymlh_get_token

Return details about the current MyMLH access token.

mymlh_refresh_token

Exchange the refresh_token for a new access token and persist it.