Uses Swagger/OpenAPI specifications to dynamically generate MCP tools, enabling AI agents to interact with any API that provides an OpenAPI specification
OpenAPI to Model Context Protocol (MCP)
The OpenAPI-MCP proxy translates OpenAPI specs into MCP tools, enabling AI agents to access external APIs without custom wrappers!
Bridge the gap between AI agents and external APIs
The OpenAPI to Model Context Protocol (MCP) proxy server bridges the gap between AI agents and external APIs by dynamically translating OpenAPI specifications into standardized MCP tools, resources, and prompts. This simplifies integration by eliminating the need for custom API wrappers.
- Repository: https://github.com/gujord/OpenAPI-MCP
If you find it useful, please give it a ⭐ on GitHub!
Key Features
- FastMCP Transport: Optimized for
stdio
, working out-of-the-box with popular LLM orchestrators. - OpenAPI Integration: Parses and registers OpenAPI operations as callable tools.
- Resource Registration: Automatically converts OpenAPI component schemas into resource objects with defined URIs.
- Prompt Generation: Generates contextual prompts based on API operations to guide LLMs in using the API.
- OAuth2 Support: Handles machine authentication via Client Credentials flow.
- JSON-RPC 2.0 Support: Fully compliant request/response structure.
- Auto Metadata: Derives tool names, summaries, and schemas from the OpenAPI specification.
- Sanitized Tool Names: Ensures compatibility with MCP name constraints.
- Flexible Parameter Parsing: Supports query strings (with a leading "?") and multiple JSON variations (including keys with dots and numeric values).
- Enhanced Parameter Handling: Automatically converts parameters to the correct data types.
- Extended Tool Metadata: Includes detailed parameter information and response schemas.
Quick Start
Installation
LLM Orchestrator Configuration
For Claude Desktop, Cursor, and Windsurf, use the snippet below and adapt the paths accordingly:
Apply this configuration to the following files:
- Cursor:
~/.cursor/mcp.json
- Windsurf:
~/.codeium/windsurf/mcp_config.json
- Claude Desktop:
~/Library/Application Support/Claude/claude_desktop_config.json
Replace
full_path_to_openapi_mcp
with your actual installation path.
Environment Configuration
Variable | Description | Required | Default |
---|---|---|---|
OPENAPI_URL | URL to the OpenAPI specification | Yes | - |
SERVER_NAME | MCP server name | No | openapi_proxy_server |
OAUTH_CLIENT_ID | OAuth client ID | No | - |
OAUTH_CLIENT_SECRET | OAuth client secret | No | - |
OAUTH_TOKEN_URL | OAuth token endpoint URL | No | - |
OAUTH_SCOPE | OAuth scope | No | api |
How It Works
- Parses OpenAPI Spec: Loads the OpenAPI specification using
httpx
andPyYAML
if needed. - Registers Operations: Extracts API operations and generates MCP-compatible tools with proper input and response schemas.
- Resource Registration: Automatically converts OpenAPI component schemas into resource objects with assigned URIs (e.g.,
/resource/{name}
). - Prompt Generation: Creates contextual prompts based on API operations to assist LLMs in understanding API usage.
- Authentication: Supports OAuth2 authentication via the Client Credentials flow.
- Parameter Handling: Converts parameters to required data types and supports flexible query string and JSON formats.
- JSON-RPC 2.0 Compliance: Ensures standard communication protocols for tool interactions.
Resources & Prompts
In addition to tools, the proxy server now automatically registers:
- Resources: Derived from OpenAPI component schemas, resource objects are registered with defined URIs (e.g.,
/resource/{name}
) for structured data handling. - Prompts: Contextual prompts are generated based on API operations to provide usage guidance to LLMs, enhancing their understanding of available endpoints.
This extended metadata improves integration by providing comprehensive API context.
Contributing
- Fork this repository.
- Create a new branch.
- Submit a pull request with a clear description of your changes.
License
If you find it useful, please give it a ⭐ on GitHub!
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
A proxy server that bridges AI agents and external APIs by dynamically translating OpenAPI specifications into standardized MCP tools, enabling seamless interaction without custom integration code.
Related MCP Servers
- AsecurityAlicenseAqualityAn MCP server that generates AI agent tools from Postman collections and requests. This server integrates with the Postman API to convert API endpoints into type-safe code that can be used with various AI frameworks.Last updated -17JavaScriptMIT License
- -securityAlicense-qualityA proxy server that enables AI assistants to run and interact with command-line applications like Expo through the Model Context Protocol (MCP), capturing logs and allowing keypress forwarding.Last updated -126TypeScriptMIT License
- -securityAlicense-qualityA command-line tool that transforms any OpenAPI service into a Model Context Protocol (MCP) server, enabling seamless integration with AI agents and tools that support the MCP specification.Last updated -902TypeScriptMIT License
- -securityFlicense-qualityA dynamic proxy that converts OpenAPI Specification (OAS) endpoints into Message Communication Protocol (MCP) tools, allowing AI agents to use existing REST APIs as if they were native MCP tools without manual implementation.Last updated -14TypeScript