postman-mcp
mcp-PostmanV3
README
Postman MCP Server
Version: v0.2.0
<a href="https://cursor.com/en-US/install-mcp?name=postman&config=eyJjb21tYW5kIjoibnBtIGkgQHBvc3RtYW52My9wb3N0bWFuLW1jcC1zZXJ2ZXIifQ%3D%3D"><img src="https://cursor.com/deeplink/mcp-install-dark.svg" alt="Add postman MCP server to Cursor" height="32" /></a>
An MCP server that provides access to the Postman API. Functionality is based on the official OpenAPI specification. For more information, see the Postman API documentation.
This project is part of the Model Context Protocol (MCP) initiative from Anthropic. For more information, visit the MCP GitHub repository and the announcement on the Anthropic blog.
Skip ahead to install instructions
[!WARNING] This project is currently under active development. Please use with caution and expect breaking changes.
[!NOTE] AI Generated Code. I used Cline v2.2.2 with Claude 3.5 Sonnet (2024-10-22). See docs/README.md for prompts and details about how this code was generated.
<a href="https://glama.ai/mcp/servers/zoig549xfd"><img width="380" height="200" src="https://glama.ai/mcp/servers/zoig549xfd/badge" alt="postman-mcp-server MCP server" /></a>
- Overview
- Features
- Installation
- Usage
- Documentation
- Rationale
- Development
- Debugging
- Other MCP Servers
- License
Overview
Postman MCP Server is a TypeScript-based MCP server that integrates with the Postman API, providing comprehensive management of Postman collections, environments, and APIs.
Features
Collections
- CRUD Operations: Create, retrieve, and update Postman collections.
- Folder Management: Organize requests into folders within collections.
- Request Management: Add and update requests within collections.
- Response Management: Manage responses associated with requests.
- Version Control: Fork, merge, and pull changes for collections.
- Comments: Add and manage comments on collections.
Environments
- Manage Environments: Create and retrieve environments for different setups.
- CRUD Operations: Full support for creating, updating, and deleting environments.
APIs
- API Management: Create, retrieve, update, and delete APIs.
- Schema Support: Manage API schemas with multi-file support.
- Tagging: Add and manage tags for APIs.
- Comments: Add and manage comments on APIs.
Authentication & Authorization
- API Key Authentication: Secure access using API keys.
- Role-Based Access Control: Manage permissions at workspace and collection levels.
- Workspace Permissions: Define permissions specific to workspaces.
Additional Features
- Private API Network: Manage elements and folders within a private API network.
- Webhooks: Create webhooks to trigger collections with custom payloads.
- Enterprise Features: Advanced role controls and SCIM support for enterprise environments.
Installation
Installing via npm (Recommended)
Install the package globally:
npm install -g @postmanv3/postman-mcp-server
Or install locally in your project:
npm install @postmanv3/postman-mcp-server
Installing via Smithery
To install Postman MCP Server for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install postman-api-server --client claude
Installing from Source
If you prefer to build from source:
-
Clone the repository:
git clone https://github.com/PostmanV3/postman-mcp-server.git cd postman-mcp-server -
Install dependencies:
npm install # or pnpm install -
Build the server:
npm run build # or pnpm run build -
Run in development mode with auto-rebuild:
npm run watch # or pnpm run watch
Usage
Setting up API Keys
-
Generate your API Key
- Visit Postman Account Settings
- Click "Generate API Key"
- Save the key securely - it won't be shown again
-
Configure the API Key
- Add the key to your environment as
POSTMAN_API_KEY - For Claude Desktop or Cline, include it in your config file (see configuration examples below)
- Never commit API keys to version control
- Add the key to your environment as
-
Verify Access
- The API key provides access to all Postman resources you have permissions for
- Test access by running a simple query (e.g., list workspaces)
[!NOTE] If you're using the Postman API collection directly, store your API key as a
postman-api-keycollection variable.
Using Claude Desktop
To use with Claude Desktop, add the server config:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%/Claude/claude_desktop_config.json
[!IMPORTANT] If you're updating this provider, Claude must be restarted to pick up API changes from the input schema (i.e. When the MCP server's ToolDefinition elements have changed). This is because Claude caches the tool definitions when it starts up.
<img width="480" alt="claude-desktop-settings" src="https://github.com/user-attachments/assets/7ea7cba2-e27e-413a-a50a-590054d51344" />
Example configuration
Recommended: Using npx (works with both global and local installations):
{
"mcpServers": {
"postman": {
"command": "npx",
"args": [
"-y",
"@postmanv3/postman-mcp-server"
],
"env": {
"POSTMAN_API_KEY": "CHANGEME"
}
}
}
}
Alternative: Direct path (if installed locally):
{
"mcpServers": {
"postman": {
"command": "node",
"args": [
"./node_modules/@postmanv3/postman-mcp-server/build/index.js"
],
"env": {
"POSTMAN_API_KEY": "CHANGEME"
}
}
}
}
Alternative: Global installation path (typical paths - find yours with npm root -g):
{
"mcpServers": {
"postman": {
"command": "node",
"args": [
"/usr/local/lib/node_modules/@postmanv3/postman-mcp-server/build/index.js"
],
"env": {
"POSTMAN_API_KEY": "CHANGEME"
}
}
}
}
[!NOTE] To find your global node_modules path, run
npm root -gand replace the path above accordingly.
[!TIP] The
npxapproach is recommended as it automatically finds the package whether it's installed globally or locally, and handles updates automatically.
Using Cline
Using the same example configuration, add the server config to your Cline MCP Servers configuration:
<img width="480" alt="cline-settings" src="https://github.com/user-attachments/assets/651ec517-9aa2-4314-84f5-bee716aa8889" />
Example configuration
Same as Claude above.
Using Zed
I'm still trying to get this to work. From the Zed docs it looks like it needs to be an extension (also this issue #21455).
Documentation
The official Postman API documentation is available in the Postman Public Workspace.
Project Overview
Postman API References & Summaries
This project leverages the Claude model and Cline extension to convert the OpenAPI specification into TypeScript code, enhancing type safety and integration within the MCP server.
This GitHub project includes API References documentation that provides detailed guidance on utilizing the Postman platform programmatically. It covers both the Collection SDK for local development and the Postman API for cloud platform integration. Key topics include authentication mechanisms, rate limits, and in-depth documentation of all API endpoints, including workspaces, collections, environments, and more. Additionally, the guide offers prerequisites and quick-start instructions to facilitate seamless API interactions.
The docs/api/summaries directory contains comprehensive Markdown summaries of the Postman API. These documents outline API endpoints, request/response formats, and implementation details essential for validating and ensuring the functionality of the MCP server. Refer to the API Summaries README for an overview of the documentation structure and implementation strategies.
Converting OpenAPI Spec to TypeScript Code with Claude
Building the MCP Server
Refer to the Handlers Documentation for detailed specifications on implementing MCP server handlers. This includes URI formats, prompt requirements, and resource handling patterns. This guide is crucial for developers working on integrating and enhancing the Postman API functionalities within the MCP server.
Rationale
The MCP wrapper for Postman tools makes sense primarily as an AI interaction layer for complex, multi-step operations where structure and safety are paramount. However, it may be overengineered for simple operations where direct CLI or API usage would suffice. The MCP wrapper provides most value when:
- Complex Operations
- Managing multiple collections
- Coordinating environments
- Generating comprehensive reports
- AI-Driven Automation
- Automated testing workflows
- API documentation maintenance
- Environment management
- Error-Sensitive Operations
- Critical API testing
- Production deployments
- Compliance checking
It provides less value for:
- Simple Operations
- Basic collection runs
- Single API calls
- Quick environment checks
- Direct CLI Usage
- Developer-driven operations
- Local testing
- Quick iterations
Development
Install dependencies:
npm install
# or
pnpm install
Build the server:
npm run build
# or
pnpm run build
For development with auto-rebuild:
npm run watch
# or
pnpm run watch
Debugging
Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, available as a package script:
npm run inspector
# or
pnpm run inspector
The Inspector will provide a URL to access debugging tools in your browser: http://localhost:5173. You will need to add the POSTMAN_API_KEY before connecting. Navigate to "Tools" to get started.
Other MCP Servers
License
This project is licensed under the MIT License. See the LICENSE file for details.
This MCP server is for research purposes only. Any data collection or monitoring will be conducted solely for security research purposes.
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。