postman-mcp
MCP server providing 41 tools to manage Postman collections, environments, mocks, API specs, and workspaces via Streamable HTTP, SSE, or stdio.
README
Postman MCP Server
A production-ready Model Context Protocol (MCP) server that provides 41 Postman API tools for managing collections, environments, mocks, API specifications, and workspaces. Supports Streamable HTTP, SSE, and stdio transports.
Features
- 📦 41 Postman API Tools - Complete Postman API automation
- 🔄 Multiple Transports - Streamable HTTP, SSE, and stdio support
- 🚀 Production Ready - Docker support with proper error handling
- 📚 Collections - Create, read, update, duplicate collections
- 🔐 Environments - Manage environment variables
- 🎭 Mocks - Create and manage mock servers
- 📝 API Specs - OpenAPI, AsyncAPI, protobuf, GraphQL support
- 🏢 Workspaces - Manage team and personal workspaces
- 🔗 Integration - Sync specs with collections
Available Tools (41)
Collections (7 tools)
createCollection- Create a collection (v2.1.0 format)getCollection- Get collection info (map/minimal/full)getCollections- List workspace collectionsputCollection- Replace collection contentsduplicateCollection- Duplicate to another workspacegetDuplicateCollectionTaskStatus- Check duplication status
Requests & Responses (3 tools)
createCollectionRequest- Create request in collectionupdateCollectionRequest- Update existing requestcreateCollectionResponse- Create request response
Environments (4 tools)
createEnvironment- Create environmentgetEnvironment- Get environment detailsgetEnvironments- List all environmentsputEnvironment- Replace environment contents
Mock Servers (5 tools)
createMock- Create mock servergetMock- Get mock server detailsgetMocks- List all mock serversupdateMock- Update mock serverpublishMock- Publish mock server (set public)
API Specifications (9 tools)
createSpec- Create API spec (OpenAPI/AsyncAPI/protobuf/GraphQL)getSpec- Get spec detailsgetAllSpecs- List workspace specsupdateSpecProperties- Update spec propertiesgetSpecDefinition- Get complete spec definitioncreateSpecFile- Create spec filegetSpecFiles- List all spec filesgetSpecFile- Get file contentsupdateSpecFile- Update spec file
Spec-Collection Integration (4 tools)
generateCollection- Generate collection from specgetSpecCollections- List spec's generated collectionsgenerateSpecFromCollection- Generate spec from collectiongetGeneratedCollectionSpecs- Get collection's generated specssyncCollectionWithSpec- Sync collection with specsyncSpecWithCollection- Sync spec with collection
Workspaces (4 tools)
createWorkspace- Create workspacegetWorkspace- Get workspace detailsgetWorkspaces- List all workspacesupdateWorkspace- Update workspace properties
Other (4 tools)
getAuthenticatedUser- Get current user infogetTaggedEntities- Get entities by tag (Enterprise)runCollection- Run collection with NewmangetEnabledTools- List enabled tools
Quick Start
Prerequisites
Required:
- Postman API Key - Get from https://postman.com/settings/me/api-keys
- Python 3.10+
Installation
cd postman-tool
# Install in editable mode
pip install -e .
# Verify installation
postman-mcp --help
Environment Variables
Required:
export POSTMAN_API_KEY="your_postman_api_key"
Windows PowerShell:
$env:POSTMAN_API_KEY = "your_postman_api_key"
Run Server
Streamable HTTP (Recommended):
postman-mcp --mode streamable-http --port 8010
SSE:
postman-mcp --mode sse --port 8010
Stdio (for MCP clients):
postman-mcp --mode stdio
Docker
# Build image
docker build -t postman-mcp .
# Run container
docker run -e POSTMAN_API_KEY=your_key -p 8010:8010 postman-mcp
MCP Client Configuration
Streamable HTTP
{
"mcpServers": {
"postman": {
"type": "streamable-http",
"url": "http://localhost:8010/mcp"
}
}
}
Stdio
{
"mcpServers": {
"postman": {
"command": "postman-mcp",
"args": ["--mode", "stdio"],
"env": {
"POSTMAN_API_KEY": "your_postman_api_key"
}
}
}
}
Usage Examples
Get Current User
await call_tool("getAuthenticatedUser", {})
List Workspaces
# List my personal workspaces
user = await call_tool("getAuthenticatedUser", {})
workspaces = await call_tool("getWorkspaces", {
"createdBy": user["user"]["id"],
"type": "personal",
"limit": 100
})
Create Collection
await call_tool("createCollection", {
"workspace": "workspace-id",
"collection": {
"info": {
"name": "My API Collection",
"description": "Collection for testing",
"schema": "https://schema.getpostman.com/json/collection/v2.1.0/collection.json"
},
"item": []
}
})
Get Collection
# Get lightweight map (default)
await call_tool("getCollection", {
"collectionId": "12345-abc123def456"
})
# Get full payload
await call_tool("getCollection", {
"collectionId": "12345-abc123def456",
"model": "full"
})
Create Mock Server
await call_tool("createMock", {
"workspace": "workspace-id",
"mock": {
"name": "My Mock Server",
"collection": "12345-abc123def456", # Collection UID
"environment": "env-id",
"private": false
}
})
Create API Specification
await call_tool("createSpec", {
"workspaceId": "workspace-id",
"name": "My API",
"type": "openapi",
"files": [
{
"path": "openapi.yaml",
"content": "openapi: 3.0.0\ninfo:\n title: My API\n version: 1.0.0"
}
]
})
Generate Collection from Spec
await call_tool("generateCollection", {
"specId": "spec-id",
"name": "Generated Collection",
"elementType": "collection",
"options": {
"requestParametersResolution": "example",
"exampleParametersResolution": "example"
}
})
Create Environment
await call_tool("createEnvironment", {
"workspace": "workspace-id",
"environment": {
"name": "Production",
"values": [
{"key": "base_url", "value": "https://api.example.com", "enabled": true},
{"key": "api_key", "value": "secret", "enabled": true, "type": "secret"}
]
}
})
Run Collection
await call_tool("runCollection", {
"collectionId": "12345-abc123def456",
"environmentId": "env-id",
"iterationCount": 1,
"requestTimeout": 30000
})
API Endpoints
When running with HTTP transports:
GET /- Server infoGET /health- Health check/mcp/*- MCP protocol endpoints (streamable-http)/sse- SSE endpoint/messages- SSE messages endpoint
Development
Run Tests
pytest
Project Structure
postman-tool/
├── postman_server.py # Main MCP server
├── tools/
│ ├── toolhandler.py # Base class
│ └── postman_tools.py # All 41 tool implementations
├── tests/
│ └── test_postman.py
├── pyproject.toml
├── Dockerfile
└── README.md
Postman API Details
Authentication
Uses X-Api-Key header with your Postman API key. Get yours at: https://postman.com/settings/me/api-keys
Base URL
https://api.getpostman.com
Collection UID Format
Many endpoints require collection UID in format: <OWNER_ID>-<COLLECTION_ID>
To get the UID:
- Use
getCollectionand read theuidfield - Construct from
{ownerId}-{collectionId}where:- For team collections:
ownerId = me.teamId(fromgetAuthenticatedUser) - For personal collections:
ownerId = me.user.id(fromgetAuthenticatedUser)
- For team collections:
Rate Limits
Postman API has rate limits. See: https://learning.postman.com/docs/developer/postman-api/postman-api-rate-limits/
Common Issues
Authentication Error
Error: Postman API error (401)
Solution:
- Verify
POSTMAN_API_KEYis set correctly - Check key is valid at https://postman.com/settings/me/api-keys
- Ensure key has required permissions
Collection UID Format
Error: Collection not found
Solution:
- Use full UID format:
12345-abc123def456 - Get UID from
getCollectionresponse - For createMock, pass collection UID, not bare ID
Workspace Required
Error: Workspace is required
Solution:
- Call
getWorkspacesto list available workspaces - Pass
workspacequery parameter - For "my workspaces", call
getAuthenticatedUserfirst
Newman Integration
The runCollection tool requires Newman (Postman's CLI runner) to be integrated programmatically. This is a placeholder that returns instructions.
Requirements
- Python 3.10+
- Postman API Key
- MCP 1.12.0+
- httpx, Starlette, Uvicorn
License
MIT License - See LICENSE file for details
Support
For issues, questions, or contributions, please visit the project repository.
Additional Resources
- Postman API Documentation: https://learning.postman.com/docs/developer/postman-api/intro-api/
- Postman Collection Format: https://schema.postman.com/collection/json/v2.1.0/draft-07/docs/index.html
- MCP Protocol: https://modelcontextprotocol.io
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。