Openapi MCP

🚀 OpenAPI to Model Context Protocol (MCP) Proxy

The OpenAPI-MCP Proxy translates OpenAPI specifications into MCP tools, enabling AI agents to access external APIs without custom wrappers!

🚀 Quick Start

📦 Installation

pip install openapi-mcp-proxy

Initialization

from openapi_mcp_proxy import MCPProxy

proxy = MCPProxy(openapi_spec_path="path/to/spec.yaml")

Get Tool Information

tools_info = proxy.list_tools()
print(tools_info)

✨ Features

• OpenAPI to MCP Conversion: Transforms OpenAPI specifications into Model Context Protocol (MCP) for use with AI toolchains like Claude and GPT.
• Tool Support: Automatically generated tools include API request, response handling, and parameter parsing functions, simplifying interaction with external systems.
• Resource Management: Generates resource objects from OpenAPI component schemas and registers them as URIs (e.g., /resource/{name}) for structured data processing.
• Prompt Integration: Automatically generates context prompts to help AI understand API functionality and usage, enhancing the interaction experience.
• OAuth2 Support: Integrates the OAuth2 client credentials flow to simplify authentication logic and improve security.

📚 Documentation

Bridging 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 converting OpenAPI specifications into standardized MCP tools, resources, and prompts.

Environment Variable Configuration

The following environment variables are required to run the OpenAPI-MCP proxy:

How It Works

sequenceDiagram
    participant LLM as LLM (Claude/GPT)
    participant MCP as OpenAPI-MCP Proxy
    participant API as External API

    Note over LLM, API: Communication Flow

    LLM->>MCP: 1. Initialize
    MCP-->>LLM: Metadata, tools, resources, and prompts

    LLM->>MCP: 2. Request tool list
    MCP-->>LLM: Detailed list of tools, including resources and prompts

    LLM->>MCP: 3. Call a tool

    alt Use OAuth2
        MCP->>API: Request OAuth2 token
        API-->>MCP: Access token
    end

    MCP->>API: 4. Execute API call with correct format
    API-->>MCP: 5. API response (JSON format)

    alt Type conversion
        MCP->>MCP: 6. Convert parameters to correct data types
    end

    MCP-->>LLM: 7. Formatted response result

    alt Dry run mode
        LLM->>MCP: Call with dry_run=true
        MCP-->>LLM: Show request information without executing the call
    end

Resources and Prompts

In addition to tools, the proxy server also automatically registers the following:

• Resources: Resource objects generated from OpenAPI component schemas and registered as defined URIs (e.g., /resource/{name}) for structured data processing.
• Prompts: Context prompts automatically generated based on API operations to help the LLM understand API functionality and usage.

This extended metadata enhances the integration experience by providing comprehensive API context information.

Contribution Guidelines

1. Fork the repository.
2. Create new features or fix issues in your branch.
3. Submit a Pull Request with a clear description of the changes.

📄 License

MIT License

If you find this project helpful, please give it a ⭐ on GitHub!