postmanlabs/postman-mcp-server

What it solves

It bridges the gap between AI agents (like Claude Code, Cursor, and GitHub Copilot) and Postman's API development ecosystem. This allows AI assistants to directly access and interact with your Postman workspaces, collections, specifications, and environments, eliminating the need to manually copy-paste API definitions or context into a chat window.

How it works

The project implements the Model Context Protocol (MCP), providing a standardized way for AI agents to communicate with Postman. It can be deployed as a remote server (via OAuth or API keys) or run locally via npm/npx. Depending on the configuration chosen (Minimal, Code, Full, or Learn), the server exposes different sets of tools to the agent, ranging from basic workspace management to generating production-ready client code and searching Postman documentation.

Who it’s for

Developers who use Postman for API design and testing and want to integrate that context into their AI-powered coding assistants or IDEs to automate API testing, synchronize code with specs, and manage collections.

Highlights

• Multiple Configuration Modes: Offers tailored toolsets for different needs, including a minimal set for basic operations, a code-focused set for client generation, and a full set with over 100 tools.
• Broad Agent Support: Compatible with a wide range of tools including Claude Code, Cursor, VS Code Copilot, and Gemini CLI.
• Flexible Deployment: Supports both remote hosted servers and local installations via Docker or npm.
• Integrated Learning: Includes a specific "Learn" mode that allows agents to search Postman's official documentation for guides and tutorials.