
Swagger MCP Server
MCP server that provides tools for exploring and testing APIs through Swagger/OpenAPI documentation.
README
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 authenticationAPI_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.
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。