MCP (Model Context Protocol)

Model Context Protocol (MCP) is a standardization of how AI assistants interact with APIs, and ReadMe is bringing this capability to your developer hub. With MCP servers, you can convert your API documentation into a structured, resource that AI assistants can understand and interact with programmatically.

Key Features

  • Custom Tooling: Define custom workflow and endpoint combinations.
    New
  • Enabled Routes: Disable endpoints you don’t want accessible in your MCP server
    New
  • OpenAPI Integration: Generate an MCP server from your existing OpenAPI specification.
  • AI assistants can connect to your MCP server to:
    • Read and understand your OpenAPI spec.
    • Execute API calls.
    • Search docs with Ask AI.
  • AI Tools Included:
    • OpenAPI Tools:
      • execute-request-body - Make API calls directly from your specification
      • get-endpoint - Pull in detailed endpoint information on demand
      • get-request-body - Access structured request parameters
      • get-response-schema - Understand what your API returns
      • list-endpoints - Browse all available API endpoints
      • list-security-schemes - Access authentication requirements
      • search-schema - Find exactly what you need in your API spec
      • get-code-snippet - Example code snippets in your preferred language to interact with your endpoint.
    • Documentation Search:
      • search-documentation - Leverage Ask AI to search your entire knowledge base

How It Works

We create a dedicated MCP server that connects to your OpenAPI specification and Ask AI functionality. This creates a bridge between your API documentation and AI assistants, making your API instantly more accessible and understandable to AI tools.

Extra Features

  • Branches By default, the MCP server is connected to the latest stable version. To chose a different branch append ?branch=<name> to the MCP url. NOTE: When you are on a branch search-documentation will not be available.
  • Private Projects To access protected projects, you will need to configure your MCP client to send a x-readme-auth header
    • Password protected: x-readme-auth should be the site password
    • Teammates only: x-readme-auth should be an api key in the form bearer <api_key>

Getting Started with MCP

Choose how you’d like to start working with MCP:

  1. Auto-Generate Your Own MCP Server: Every ReadMe project automatically includes a fully configured MCP server. Simply enable MCP to connect your API documentation to AI tools.
  2. Use ReadMe’s MCP Server: With ReadMe’s MCP server, you can do everything you normally would in ReadMe, like adding and editing pages, directly through our API.