OpenAPI to MCP
A Go tool for converting OpenAPI specifications into MCP tools.
openapi-to-mcp
简体中文
openapi-to-mcp is a Go implementation for converting OpenAPI specifications into MCP (Model Context Protocol) tools.
Overview
This project aims to provide a flexible framework that allows developers to automatically expose existing OpenAPI interfaces as MCP tools through simple configuration, enabling AI models to call these interfaces directly.
Features
- OpenAPI to MCP Tool Conversion: Automatically parses OpenAPI specifications and creates corresponding MCP tools based on the definitions.
- Multiple Transport Support: Supports
stdio(Standard I/O),sse(Server-Sent Events), andstream(HTTP Stream) as transport protocols for MCP communication. - State Tracking & Authentication: Supports cookie-based state tracking and JWT (JSON Web Token) handling.
- Rate Limiting: Built-in rate limiting to prevent high-frequency calls to the Large Language Model (LLM).
- Environment Variable Configuration: Flexible configuration via
.envfile or system environment variables. - Stricter MCPTool Definition: Defines tools more rigorously for better usability by LLMs.
Installation
You can install openapi-to-mcp in two ways.
Option 1: Using go install (Recommended)
The easiest way to install and run the tool is using go install. After installation, the openapi-to-mcp command will be available in your shell.
go install github.com/constellation39/openapi-to-mcp@latest
Option 2: From Source
If you want to modify the code or contribute, clone the repository and build the project.
git clone https://github.com/constellation39/openapi-to-mcp cd openapi-to-mcp go build .
The executable will be ./openapi-to-mcp
Usage
Follow these steps to configure and run the tool.
Step 1: Configure Environment Variables
Create a .env file in the project's root directory or set the following environment variables in your system. You can start by copying the example file:
On Windows
copy .env.example .env
On macOS/Linux
cp .env.example .env
Then, edit the .env file with your desired configuration:
MCP transport: stdio, sse, stream (default: stdio)
MCP_TRANSPORT="stdio"
MCP BASE URL (used for sse and stream modes)
MCP_BASE_URL="http://localhost:8080"
OpenAPI specification file path (can be a local file path or a URL)
OPENAPI_SRC="./example/openapi.yaml"
Base URL for API requests
OPENAPI_BASE_URL=
Extra HTTP headers (JSON format), e.g., '{"X-API-Key": "your-api-key"}'
EXTRA_HEADERS='{"X-API-Key": "your-api-key"}'
Use cookies (true/false, default: true)
USE_COOKIE=true
Output logs to standard output (true/false, default: false)
LOG_OUTPUT=false
Rate limit: allowed requests per second (default: 1)
RATE_LIMIT_PER_SECOND=1
Authorization header, e.g., "Basic xxxx"
AUTHORIZATION_HEADERS="Basic xxxx"
Step 2: Run the Application
Execute the application from your terminal.
If you installed using go install:
openapi-to-mcp
If you built from source:
./openapi-to-mcp
For development, you can also run directly using go run:
go run main.go
Step 3: Connect with an MCP Client
Configure your MCP client to connect to openapi-to-mcp based on the MCP_TRANSPORT you've chosen.
- Stdio Mode (Default): If MCP_TRANSPORT is stdio, configure your client to execute the command directly. The env block can be used to pass or override environment variables.
{ "mcpServers": { "openapi-to-mcp": { "command": "openapi-to-mcp", "env": { "MCP_BASE_URL": "http://localhost:8080", "OPENAPI_SRC": "./example/openapi.yaml" } } } }
Note: If running from source, you might set "command" to "./openapi-to-mcp" or use go run main.go.
- SSE Mode: If MCP_TRANSPORT is sse, the server will start on the MCP_BASE_URL. Configure your client to connect to the /sse endpoint.
{ "mcpServers": { "openapi-to-mcp": { "url": "http://localhost:8080/sse" } } }
- Stream Mode: If MCP_TRANSPORT is stream, the server will start on the MCP_BASE_URL. Configure your client to connect to the /stream endpoint.
{ "mcpServers": { "openapi-to-mcp": { "url": "http://localhost:8080/stream" } } }
Please ensure http://localhost:8080 matches the MCPBASEURL in your configuration.
Project Structure
.github/
core/
├── openapi.go # OpenAPI specification parsing and tool generation logic
├── session/ # Session management
├── middleware.go # MCP middleware definitions
└── utils.go # Common utility functions
example/
├── openapi.yaml # Example OpenAPI specification file
main.go # Main application entry point
README.md
README_zh.md
go.mod
go.sum
Contributing
Contributions are welcome! Feel free to submit a Pull Request or report an Issue.
License
This project is released under the MIT License.
相關伺服器
Scout Monitoring MCP
贊助Put performance and error data directly in the hands of your AI assistant.
Alpha Vantage MCP Server
贊助Access financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
MCP Jupiter
Access Jupiter's swap API on the Solana blockchain.
CodeRabbit
Interact with CodeRabbit AI reviews on GitHub pull requests.
Proteus Workflow Engine
A modern, extensible multi-agent workflow engine with real-time monitoring and a web visualization interface.
Software Planning Tool
A tool for structured software development planning, helping to break down projects into tasks and track progress.
SD Elements MCP Server
Provides SD Elements API integration for security and compliance management.
Projet MCP Server-Client
An implementation of the Model Context Protocol (MCP) for communication between AI models and external tools, featuring server and client examples in Python and Spring Boot.
MCP-Portainer Bridge
Manage Docker containers through the Portainer API.
ThousandEyes MCP Server
ThousandEyes Network Performance Monitoring MCP Server
Roo Activity Logger
Automatically logs AI coding assistant activities, such as command executions and code generation, into searchable JSON files.
Swiftzilla
The only RAG API built for Apple Development. Give your AI instant access to 100,000+ pages of official docs, recipes, and evolution proposals.