Swagger MCP Server

A Model Context Protocol (MCP) server that provides tools for exploring and testing APIs through Swagger/OpenAPI documentation. This server automatically detects configuration files from multiple IDEs and provides comprehensive API interaction capabilities.

Features

🔍 Fetch and parse Swagger/OpenAPI documentation from any URL
🧪 Test API endpoints directly through the MCP interface
📊 Explore API schemas and understand data structures
🔧 Multi-IDE support - automatically detects config from VS Code, Cursor, Windsurf, and more
🌐 Flexible authentication - supports API keys, basic auth, and bearer tokens
⚡ Auto-discovery - can find documentation URLs automatically

Configuration

IDE Setup

Create an MCP configuration file in your IDE's configuration directory:

VS Code: ~/.vscode/mcp.json or .vscode/mcp.json (in your project)
Cursor: ~/.cursor/mcp.json or .cursor/mcp.json (in your project)
Windsurf: ~/.windsurf/mcp.json or .windsurf/mcp.json (in your project)
Any IDE: mcp.json (in your project root) or .mcp/config.json

Authentication Options

Option 1: Using API Key

"swagger-mcp": {
  "command": "npx",
  "args": [
    "-y",
    "swagger-mcp@latest"
  ],
  "env": {
    "API_BASE_URL": "https://api.example.com",
    "API_DOCS_URL": "https://api.example.com/swagger.json",
    "API_KEY": "your-api-key-here"
  }
}

Option 2: Using Username and Password

"swagger-mcp": {
  "command": "npx",
  "args": [
    "-y", 
    "swagger-mcp@latest"
  ],
  "env": {
    "API_BASE_URL": "https://api.example.com",
    "API_DOCS_URL": "https://api.example.com/swagger.json",
    "API_USERNAME": "your-username",
    "API_PASSWORD": "your-password"
  }
}

Configuration Options

API_BASE_URL - Base URL for your API (e.g., https://api.example.com) [Required]
API_DOCS_URL - Direct URL to Swagger/OpenAPI JSON/YAML (optional, will be auto-discovered)
API_KEY - API key for authentication (used as Bearer token)
API_USERNAME - Username for basic authentication
API_PASSWORD - Password for basic authentication

Authentication Flow

The server intelligently handles authentication:

For API requests: Uses API_KEY as Bearer token, falls back to Basic auth
For authentication endpoints: Auto-injects username/password credentials
Token management: Automatically stores and reuses tokens from login responses
Auto-refresh: Attempts to refresh tokens on 401 Unauthorized responses

Available Tools

`fetch_swagger_info`

Fetches and parses Swagger/OpenAPI documentation from a given URL to discover available API endpoints.

`list_endpoints`

Lists all available API endpoints after fetching Swagger documentation, showing methods, paths, and summaries.

`get_endpoint_details`

Gets detailed information about a specific API endpoint including parameters, request/response schemas, and examples.

`execute_api_request`

Executes an API request to a specific endpoint with authentication, parameters, headers, and body handling.

`validate_api_response`

Validates an API response against the schema definitions from Swagger documentation to ensure compliance.

Usage Examples

Once configured, you can use the MCP server in your AI-powered editor to:

Explore APIs: "Show me the available endpoints in this API"
Test endpoints: "Test the POST /users endpoint with this data"
Understand schemas: "Explain the User model structure"
Debug API calls: "Help me troubleshoot this API request"
Validate responses: "Check if this response matches the API schema"

Supported IDEs

The server automatically detects configuration files from:

VS Code (.vscode/mcp.json)
Cursor (.cursor/mcp.json)
Windsurf (.windsurf/mcp.json)
Root directory (mcp.json)
Alternative location (.mcp/config.json)

Development

# Clone the repository
git clone https://github.com/amrsa1/SwaggerMCP.git
cd SwaggerMCP

# Install dependencies
npm install

# Run in development mode
npm run dev

# Build for production
npm run build

License

MIT License - see LICENSE file for details.

Contributing

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