apidog-mcp

apidog-mcp

An MCP server for Apidog that enables listing projects, inspecting API trees, creating modules and folders, and adding or updating requests without manual configuration.

Category
访问服务器

README

apidog-mcp

apidog-mcp is a stdio MCP server for Apidog. It lets an MCP client such as opencode list projects, inspect API trees, create modules and folders, and add or update requests without configuring every Apidog project by hand.

Quick start

Use the published package with npx from your MCP client configuration.

{
  "mcp": {
    "apidog": {
      "type": "local",
      "command": ["npx", "-y", "@byixlam/apidog-mcp"],
      "environment": {
        "APIDOG_ACCESS_TOKEN": "adgp_xxx",
        "APIDOG_TEAM_ID": "679680"
      }
    }
  }
}

APIDOG_ACCESS_TOKEN is required for Apidog API calls. The server still starts without it, but tools will return a clear setup error until the token is configured and the MCP client is restarted.

Optional environment variables:

  • APIDOG_TEAM_ID: default team for apidog_list_team_projects.
  • APIDOG_LOCALE: defaults to en-US.
  • APIDOG_CLIENT_VERSION: defaults to 2.8.34.

Configure a default project

Most tools accept projectId. If you do not want to pass it every time, run apidog_configure_project once in the workspace:

{
  "projectId": 1312991,
  "branchId": 0
}

This writes apidog.json and adds it to .gitignore.

Tools

Setup and discovery

  • apidog_check_auth
  • apidog_configure_project
  • apidog_list_user_projects
  • apidog_list_user_teams
  • apidog_list_team_projects

Project navigation

  • apidog_list_modules
  • apidog_list_environments
  • apidog_list_api_tree
  • apidog_list_api_detail_folders
  • apidog_list_api_responses
  • apidog_list_test_case_categories
  • apidog_list_test_case_tags
  • apidog_list_runners
  • apidog_get_http_api

Modules

  • apidog_create_module
  • apidog_update_module
  • apidog_rename_module
  • apidog_delete_module

Delete tools require confirmDelete: true.

Folders

  • apidog_create_api_detail_folder
  • apidog_rename_api_detail_folder
  • apidog_delete_api_detail_folder

Delete tools require confirmDelete: true.

Apidog-native requests

These tools work directly with Apidog HTTP API details.

  • apidog_add_request
  • apidog_update_request
  • apidog_delete_request
  • apidog_create_api_detail
  • apidog_update_api_detail
  • apidog_delete_api_detail

Use apidog_add_request and apidog_update_request for normal request creation and editing. Use the raw api_detail tools when you need full control over Apidog-specific fields.

When a URL contains path parameters such as /users/{id}/posts/{postId}, the request tools automatically create path parameters for id and postId unless they already exist.

OpenAPI import/export

  • apidog_export_openapi
  • apidog_get_openapi
  • apidog_validate_openapi
  • apidog_import_openapi
  • apidog_add_endpoint
  • apidog_update_endpoint
  • apidog_upsert_endpoint
  • apidog_delete_endpoint
  • apidog_patch_openapi

The endpoint tools use an export-modify-validate-import flow. Before importing anything, the server validates the OpenAPI document. apidog_delete_endpoint requires both confirmDelete: true and deleteUnmatchedResources: true.

Common examples

Create a folder

{
  "moduleId": 1312066,
  "name": "Billing",
  "parentId": 0,
  "type": "http"
}

Add a request inside a folder

{
  "moduleId": 1312066,
  "folderId": 9034034,
  "name": "Get invoice",
  "method": "get",
  "path": "/invoices/{invoiceId}",
  "status": "developing",
  "responses": [
    {
      "code": 200,
      "name": "Success",
      "contentType": "json",
      "jsonSchema": {
        "type": "object",
        "properties": {
          "invoiceId": { "type": "string" }
        }
      },
      "itemSchema": {}
    }
  ]
}

Add a folder placeholder request

Apidog allows a request with an empty path inside a folder. This is useful for drafts.

{
  "moduleId": 1312066,
  "folderId": 9034034,
  "name": "Draft request",
  "method": "get",
  "path": ""
}

Update an endpoint through OpenAPI

{
  "moduleId": 1312066,
  "path": "/invoices/{invoiceId}",
  "method": "get",
  "patch": {
    "summary": "Get invoice by ID",
    "responses": {
      "200": {
        "description": "Invoice returned"
      }
    }
  }
}

Delete a request

{
  "apiDetailId": 37814603,
  "confirmDelete": true
}

How it works

The server talks to https://api.apidog.com and sends the access token as a Bearer token. Project-scoped calls include the project ID and branch ID headers when available.

There are two editing paths:

  1. Direct Apidog tools for modules, folders, and Apidog-native request details.
  2. OpenAPI tools for contract-style endpoint changes using Apidog import/export.

Use direct request tools when you care about Apidog-specific fields. Use OpenAPI tools when you want to make portable API contract changes.

Local development

Install dependencies and build:

npm install
npm run build

Run checks:

npm run typecheck
npm pack --dry-run

Start locally:

APIDOG_ACCESS_TOKEN=adgp_xxx node dist/index.js

The server uses stdio, so it does not print a startup banner.

Testing safely

When testing write tools against a real Apidog project:

  1. Use a disposable project or copy.
  2. Create a temporary module with a unique name.
  3. Create folders and requests only inside that module.
  4. Delete the temporary module when finished.
  5. List modules and the API tree to confirm cleanup.

Avoid testing delete tools against existing project resources.

Notes

This project uses both documented Apidog import/export APIs and endpoints used by the Apidog web app. Those web-app endpoints can change without notice.

Keep destructive tools behind confirmDelete: true, and do not store session cookies or other browser session values in this repository.

Donations

USDT BEP20: 0x589aC54204D9BfBe7E455E09b5fcbD4231C0F55b

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