swagger-doc-explorer-mcp

swagger-doc-explorer-mcp

Enables progressive exploration of Swagger/OpenAPI API documentation. Supports loading multiple specs, browsing endpoints and schemas, and searching across documentation.

Category
访问服务器

README

Swagger Doc Explorer MCP

MCP server for progressive exploration of Swagger/OpenAPI documentation. Supports stdio (local) and HTTP (remote) transports.

Each loaded spec is identified by a unique name ({title} v{version}) assigned at load time. All downstream tools reference the spec by name, so you can load and inspect multiple specs simultaneously.

Tools

Tool Description
swagger_load_spec Load spec from a remote URL. url (required), auth_header (optional). Returns the assigned spec_name
swagger_load_local_spec Load spec from a local JSON file. file_path (required). Returns the assigned spec_name
swagger_list_loaded List all specs currently loaded (name, title, version, source)
swagger_remove_spec Remove a loaded spec from memory. spec_name (required)
swagger_get_info Get API title, version, description, server URL. spec_name (required)
swagger_list_tags List all tags/groups with endpoint counts. spec_name (required)
swagger_list_paths List endpoints, optionally filtered by tag. Supports limit (default 50) and offset pagination. spec_name (required)
swagger_get_endpoint Drill into a specific endpoint: path, method. spec_name (required)
swagger_get_endpoint_full Drill into endpoint with all $ref recursively resolved. Single-call completeness
swagger_list_schemas List all data models/schemas. Supports limit and offset. spec_name (required)
swagger_get_schema Get full details of a specific schema: schema_name. spec_name (required)
swagger_search Full-text search across endpoints and schemas: query. spec_name (required)

Installation

npm install -g swagger-doc-explorer-mcp

Or use directly with npx (no install needed):

npx -y swagger-doc-explorer-mcp

Usage

stdio (default)

npx -y swagger-doc-explorer-mcp

HTTP

SWAGGER_HTTP_PORT=3000 npx -y swagger-doc-explorer-mcp

The HTTP server accepts POST requests at / following the MCP Streamable HTTP protocol. Clients must include Accept: application/json, text/event-stream and use the mcp-session-id header for session affinity:

# Initialize session
curl -X POST http://localhost:3000/ \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"my-client","version":"1.0"}}}'

# Copy the mcp-session-id from response headers, then:
curl -X POST http://localhost:3000/ \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "mcp-session-id: <session-id>" \
  -d '{"jsonrpc":"2.0","method":"notifications/initialized"}'

curl -X POST http://localhost:3000/ \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "mcp-session-id: <session-id>" \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"swagger_load_spec","arguments":{"url":"https://petstore.swagger.io/v2/swagger.json"}}}'

Development

npm run dev          # stdio mode with hot reload
npm run dev:http     # HTTP mode on port 3000 with hot reload

MCP client configuration

Use npx so the client auto-installs the package:

Claude Desktop (claude_desktop_config.json):

{
  "mcpServers": {
    "swagger-doc-explorer": {
      "command": "npx",
      "args": ["-y", "swagger-doc-explorer-mcp"]
    }
  }
}

VS Code (.vscode/mcp.json):

{
  "servers": {
    "swagger-doc-explorer": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "swagger-doc-explorer-mcp"]
    }
  }
}

If installed globally, use swagger-doc-explorer-mcp directly as the command.

opencode (~/.config/opencode/opencode.json):

{
  "mcp": {
    "swagger-doc-explorer": {
      "type": "local",
      "command": ["npx", "-y", "swagger-doc-explorer-mcp"],
      "enabled": true
    }
  }
}

Environment Variables

Variable Description
SWAGGER_HTTP_PORT Set to enable HTTP mode (e.g. 3000). Omit for stdio mode

Progressive Exploration Workflow

1. swagger_load_spec / swagger_load_local_spec  → load the API spec (returns spec_name)
2. swagger_get_info(spec_name="...")             → overview of the API
3. swagger_list_tags(spec_name="...")            → see available groups
4. swagger_list_paths(spec_name="...", tag="users") → browse endpoints in a group
5. swagger_get_endpoint(spec_name="...", path, method) → drill into endpoint details
6. swagger_list_schemas(spec_name="...")         → see available data models
7. swagger_get_schema(spec_name="...", schema_name) → inspect a model
8. swagger_search(spec_name="...", query)        → find anything across the spec

Multiple specs can be loaded and queried simultaneously — each identified by its unique spec_name.

Compatibility

  • OpenAPI v3.x (openapi field)
  • Swagger v2 (swagger field)
  • JSON format only (YAML is not supported)
  • Remote URLs and local file paths

Build & Test

npm run build
npm test

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选