Openapi MCP Proxy

OpenAPI MCP Server is a command-line tool that can convert any OpenAPI service into a server compliant with the MCP protocol, enabling seamless integration and use of these services by AI agents and tools supporting the MCP specification.

Developer tools Artificial intelligence chatbots #API conversion #AI integration #Protocol bridging #Tool proxy .TypeScript

rating : 2.5 points

downloads : 20

update time : 2025-04-28

What is OpenAPI MCP Server?

OpenAPI MCP Server is a middleware tool that automatically converts existing OpenAPI (REST API) specifications into servers compliant with the Model Context Protocol (MCP) standard. This allows any existing REST API to be immediately accessed and used by AI agents supporting MCP without modifying the original service.

How to use OpenAPI MCP Server?

Simply provide the OpenAPI specification file and the target API address, and the tool will automatically generate the MCP server and handle protocol conversion. You can interact with the converted API through an MCP-compatible client (such as VS Code Copilot) or a dedicated inspector.

Applicable scenarios

When you want an AI agent (such as GitHub Copilot) to directly use your existing REST API; when you need to unify the interaction methods of different APIs for AI system integration; quickly add AI agent support capabilities to existing services.

Main features

Automatic OpenAPI conversionAutomatically parse the OpenAPI specification and generate corresponding MCP service endpoints

Protocol conversionConvert MCP requests to REST calls in real-time and convert responses back to the MCP format

Service discoveryExpose all available API operations through the standard MCP discovery endpoint

VS Code integrationSupport direct integration with the Agent mode of VS Code Copilot

Advantages and limitations

Advantages

Zero-code transformation - Add AI support without modifying existing APIs

Standardized interface - Provide a consistent AI interaction method through MCP

Wide compatibility - Support any REST service compliant with the OpenAPI specification

Developer-friendly - Provide detailed testing and debugging tools

Limitations

Early version - Features are still being continuously improved

Performance overhead - As a proxy layer, it will add a small amount of latency

Protocol limitation - Only support REST operations defined by OpenAPI

Authentication - Complex API authentication schemes may require additional configuration

How to use

Install the tool

Install openapi-mcp-proxy globally via npm

Prepare the OpenAPI file

Ensure you have the OpenAPI specification file (in YAML or JSON format) for the API

Start the MCP server

Run the command to start the conversion proxy, specifying the specification file and the target API address

Test the connection

Use the MCP inspector or VS Code to verify that the service is running properly

Usage examples

Pet store API integrationConvert the Swagger Petstore API into an MCP service for AI agents to use

Empowering enterprise internal APIsOpen the company's internal employee management system API for Copilot to use

Frequently Asked Questions

Does the MCP server need to run continuously?

Which authentication methods are supported?

How to verify that the conversion is correct?

Does it support WebSocket or other non-REST protocols?

Related resources

Project roadmap

Understand the project development plan and future features

Official MCP documentation

Model Context Protocol specification and usage guide

OpenAPI specification

Official OpenAPI/Swagger specification documentation

Demo video

Demonstration of the tool in actual operation

🚀 OpenAPI MCP Server

A command-line tool that can convert any OpenAPI service into a Model Context Protocol (MCP) server, enabling seamless integration with AI agents and tools supporting the MCP specification.

⚠️ Important Note

This tool is still in the early stages of development. The roadmap is here

🚀 Quick Start

What is Model Context Protocol (MCP)?

Model Context Protocol (MCP) is a standardized protocol for the interaction between AI agents and external tools and services. It provides a unified way to discover, understand, and utilize various capabilities, making it easier to build AI-driven applications that can interact with multiple services.

What is OpenAPI?

OpenAPI (formerly known as Swagger) is a standard for describing machine-readable interface files, used to define, generate, consume, and visualize RESTful web services. It allows humans and computers to discover and understand the functionality of services without accessing the source code or additional documentation.

About the Project

This middleware acts as a bridge between OpenAPI services and MCP clients:

Accepts an OpenAPI specification file as input.
Automatically generates a server that complies with the MCP standard.
Proxies requests to the actual OpenAPI service.
Converts between MCP and REST conventions.

This enables any existing REST API with an OpenAPI specification to be immediately accessible to AI agents that understand MCP, without modifying the original service.

📦 Installation

npm install -g openapi-mcp-proxy

💻 Usage Examples

Basic Usage

npx openapi-mcp-proxy --spec ./path/to/openapi.yml --target http://your-api.com --port 3000

Options

-s, --spec <path> (Required): Path to the OpenAPI specification file (Environment variable: OMP_OPENAPI_SPEC_FILE).
-t, --target <url> (Optional): Target API service URL (Default: "http://localhost:8080", Environment variable: OMP_TARGET_BASE_URL).
-p, --port <number> (Optional): Port on which the MCP server runs (Default: "3000", Environment variable: PORT).
-V, --version: Display the version number.
-h, --help: Display help information.

Advanced Usage

Start the MCP server for the Petstore API:

npx openapi-mcp-proxy --spec resources/petstore.openapi.yml --target https://petstore.swagger.io/v2

Testing

MCP Check Tool

To test your MCP server, do the following:

Install the check tool: npm install -g @merossai/mcpx.
Start the check tool: mcpx <your_api_key> http://localhost:3000.

Configuration in VS Code

In VS Code, you can press Ctrl+Shift+P to open the command palette:

Type and select "Remote-Containers: Start Server".
Configure the proxy settings:
- Proxy URL: http://localhost:3000.
- Directly connect to the target port: true.