@reapi/mcp 开放API

什么是 MCP OpenAPI？

MCP OpenAPI 是一个开源项目，提供了 MCP（多通道协议）服务器的规范。它允许开发者以标准化的格式定义和记录 API，从而更容易地集成不同的系统和服务。MCP OpenAPI 规范旨在增强互操作性，并简化依赖多通道通信的应用程序的开发过程。

MCP OpenAPI 的特点

• 标准化的 API 文档：MCP OpenAPI 提供了一种清晰简洁的方式来记录 API，确保开发者能够轻松理解如何与各种服务进行交互。
• 互操作性：通过遵循 OpenAPI 规范，MCP OpenAPI 促进了不同系统之间的无缝通信，无论底层技术如何。
• 开源：作为一个开源项目，MCP OpenAPI 鼓励社区贡献，允许开发者增强和扩展其功能。
• 用户友好的界面：该项目为开发者提供了一个用户友好的界面，用于创建、管理和测试 API，减少了与 API 开发相关的学习曲线。

如何使用 MCP OpenAPI

1. 安装：首先从 GitHub 克隆 MCP OpenAPI 仓库。您可以使用以下命令：

   git clone https://github.com/ReAPI-com/mcp-openapi.git
   

2. 定义您的 API：创建一个 OpenAPI 规范文件（通常为 YAML 或 JSON 格式），概述您的 API 的端点、请求/响应格式和认证方法。
3. 生成文档：使用 Swagger UI 或 ReDoc 等工具从您的 OpenAPI 规范生成交互式文档，使开发者更容易理解如何使用您的 API。
4. 测试：利用 Postman 或 curl 等工具测试您的 API 端点，确保它们按预期工作。
5. 部署：一旦您的 API 准备就绪，将其部署到您的服务器或云平台，使其可供用户和其他应用程序访问。

常见问题解答

问：MCP OpenAPI 的目的是什么？

答：MCP OpenAPI 的目的是提供一种标准化的方式来定义和记录 API，促进不同系统之间更好的通信和集成。

问：MCP OpenAPI 是免费使用的吗？

答：是的，MCP OpenAPI 是一个开源项目，这意味着它可以在其许可证条款下免费使用、修改和分发。

问：我可以为 MCP OpenAPI 项目做贡献吗？

答：当然可以！欢迎贡献。您可以通过报告问题、提交功能请求，甚至创建包含代码改进的拉取请求来贡献。

问：我在哪里可以找到有关 MCP OpenAPI 的更多信息？

答：您可以在 GitHub 的 官方仓库 上找到更多信息，包括文档和示例。

@reapi/mcp-openapi

A Model Context Protocol (MCP) server that loads and serves multiple OpenAPI specifications to enable LLM-powered IDE integrations. This server acts as a bridge between your OpenAPI specifications and LLM-powered development tools like Cursor and other code editors.

Features

• Loads multiple OpenAPI specifications from a directory
• Exposes API operations and schemas through MCP protocol
• Enables LLMs to understand and work with your APIs directly in your IDE
• Supports dereferenced schemas for complete API context
• Maintains a catalog of all available APIs

Powered by ReAPI

This open-source MCP server is sponsored by ReAPI, a next-generation API platform that simplifies API design and testing. While this server provides local OpenAPI integration for development, ReAPI offers two powerful modules:

🎨 API CMS

• Design APIs using an intuitive no-code editor
• Generate and publish OpenAPI specifications automatically
• Collaborate with team members in real-time
• Version control and change management

🧪 API Testing

• The most developer-friendly no-code API testing solution
• Create and manage test cases with an intuitive interface
• Powerful assertion and validation capabilities
• Serverless cloud test executor
• Perfect for both QA teams and developers
• CI/CD integration ready

Try ReAPI for free at reapi.com and experience the future of API development.

Cursor Configuration

To integrate the MCP OpenAPI server with Cursor IDE, you have two options for configuration locations:

Option 1: Project-specific Configuration (Recommended)

Create a .cursor/mcp.json file in your project directory. This option is recommended as it allows you to maintain different sets of specs for different projects

{
  "mcpServers": {
    "@reapi/mcp-openapi": {
      "command": "npx",
      "args": ["-y", "@reapi/mcp-openapi@latest", "--dir", "./specs"],
      "env": {}
    }
  }
}

Tip: Using a relative path like ./specs makes the configuration portable and easier to share across team members.

Note: We recommend using @latest tag as we frequently update the server with new features and improvements.

Important: Project-specific configuration helps manage LLM context limits. When all specifications are placed in a single folder, the combined metadata could exceed the LLM's context window, leading to errors. Organizing specs by project keeps the context size manageable.

Option 2: Global Configuration

Create or edit ~/.cursor/mcp.json in your home directory to make the server available across all projects:

{
  "mcpServers": {
    "@reapi/mcp-openapi": {
      "command": "npx",
      "args": ["-y", "@reapi/mcp-openapi@latest", "--dir", "/path/to/your/specs"],
      "env": {}
    }
  }
}

Enable in Cursor Settings

After adding the configuration:

1. Open Cursor IDE
2. Go to Settings > Cursor Settings > MCP
3. Enable the @reapi/mcp-openapi server
4. Click the refresh icon next to the server to apply changes

Note: By default, Cursor requires confirmation for each MCP tool execution. If you want to allow automatic execution without confirmation, you can enable Yolo mode in Cursor settings.

The server is now ready to use. When you add new OpenAPI specifications to your directory, you can refresh the catalog by:

1. Opening Cursor's chat panel
2. Typing one of these prompts:

   "Please refresh the API catalog"
   "Reload the OpenAPI specifications"
   

OpenAPI Specification Requirements

1. Place your OpenAPI 3.x specifications in the target directory:

   • Supports both JSON and YAML formats
   • Files should have .json, .yaml, or .yml extensions
   • Scanner will automatically discover and process all specification files

2. Specification ID Configuration:

   • By default, the filename (without extension) is used as the specification ID
   • To specify a custom ID, add x-spec-id in the OpenAPI info object:

   openapi: 3.0.0
   info:
     title: My API
     version: 1.0.0
     x-spec-id: my-custom-api-id  # Custom specification ID
   

   Important: Setting a custom x-spec-id is crucial when working with multiple specifications that have:

   • Similar or identical endpoint paths
   • Same schema names
   • Overlapping operation IDs

   The spec ID helps distinguish between these similar resources and prevents naming conflicts. For example:

   # user-service.yaml
   info:
     x-spec-id: user-service
   paths:
     /users:
       get: ...
   
   # admin-service.yaml
   info:
     x-spec-id: admin-service
   paths:
     /users:
       get: ...
   

   Now you can reference these endpoints specifically as user-service/users and admin-service/users

How It Works

1. The server scans the specified directory for OpenAPI specification files
2. It processes and dereferences the specifications for complete context
3. Creates and maintains a catalog of all API operations and schemas
4. Exposes this information through the MCP protocol
5. IDE integrations can then use this information to:
   • Provide API context to LLMs
   • Enable intelligent code completion
   • Assist in API integration
   • Generate API-aware code snippets

Tools

1. refresh-api-catalog

   • Refresh the API catalog
   • Returns: Success message when catalog is refreshed

2. get-api-catalog

   • Get the API catalog, the catalog contains metadata about all openapi specifications, their operations and schemas
   • Returns: Complete API catalog with all specifications, operations, and schemas

3. search-api-operations

   • Search for operations across specifications
   • Inputs:
     • query (string): Search query
     • specId (optional string): Specific API specification ID to search within
   • Returns: Matching operations from the API catalog

4. search-api-schemas

   • Search for schemas across specifications
   • Inputs:
     • query (string): Search query
     • specId (optional string): Specific API specification ID to search
   • Returns: Matching schemas from the API catalog

5. load-api-operation-by-operationId

   • Load an operation by operationId
   • Inputs:
     • specId (string): API specification ID
     • operationId (string): Operation ID to load
   • Returns: Complete operation details

6. load-api-operation-by-path-and-method

   • Load an operation by path and method
   • Inputs:
     • specId (string): API specification ID
     • path (string): API endpoint path
     • method (string): HTTP method
   • Returns: Complete operation details

7. load-api-schema-by-schemaName

   • Load a schema by schemaName
   • Inputs:
     • specId (string): API specification ID
     • schemaName (string): Name of the schema to load
   • Returns: Complete schema details

Roadmap

1. Semantic Search

   • Enable natural language queries for API operations and schemas
   • Improve search accuracy with semantic understanding

2. Remote Specs Sync

   • Support syncing OpenAPI specifications from remote sources

3. Code Templates

   • Expose code templates through MCP protocol
   • Provide reference patterns for LLM code generation

4. Community Contributions

   • Submit feature requests and bug reports
   • Contribute to improve the server

Example Prompts in Cursor

Here are some example prompts you can use in Cursor IDE to interact with your APIs:

1. Explore Available APIs

   "Show me all available APIs in the catalog with their operations"
   "List all API specifications and their endpoints"
   

2. API Operation Details

   "Show me the details of the create pet API endpoint"
   "What are the required parameters for creating a new pet?"
   "Explain the response schema for the pet creation endpoint"
   

3. Schema and Mock Data

   "Generate mock data for the Pet schema"
   "Create a valid request payload for the create pet endpoint"
   "Show me examples of valid pet objects based on the schema"
   

4. Code Generation

   "Generate an Axios client for the create pet API"
   "Create a TypeScript interface for the Pet schema"
   "Write a React hook that calls the create pet endpoint"
   

5. API Integration Assistance

   "Help me implement error handling for the pet API endpoints"
   "Generate unit tests for the pet API client"
   "Create a service class that encapsulates all pet-related API calls"
   

6. Documentation and Usage

   "Show me example usage of the pet API with curl"
   "Generate JSDoc comments for the pet API client methods"
   "Create a README section explaining the pet API integration"
   

7. Validation and Types

   "Generate Zod validation schema for the Pet model"
   "Create TypeScript types for all pet-related API responses"
   "Help me implement request payload validation for the pet endpoints"
   

8. API Search and Discovery

   "Find all endpoints related to pet management"
   "Show me all APIs that accept file uploads"
   "List all endpoints that return paginated responses"
   

These prompts demonstrate how to leverage the MCP server's capabilities for API development. Feel free to adapt them to your specific needs or combine them for more complex tasks.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.