MCP Server

Give your AI agents trusted access to the full Postman platform. Connect any MCP-compatible agent to 100+ Postman tools—collections, specs, environments, workspaces and more.

Get StartedLearn More
Postman MCP Server connecting AI agents to API workflows

Connect top coding agents

Claude Code

Codex

Gemini CLI

VS Code

Antigravity*

Cursor

Windsurf

Kiro

Claude*

*Local only.

Toolsets

Minimal

Essential tools for core Postman operations.

mcp.postman.com/minimal

Code

Search APIs and generate idiomatic client code.

mcp.postman.com/code

Full

See all 100+ Postman API tools.

mcp.postman.com/mcp

Use AI agents to speed up API development

With Postman's MCP server, you can prompt AI to automate work across your Postman collections, environments, workspaces, and more.

Try it out with these sample prompts:

  • "Set the base_url variable in my Postman Development environment to http://localhost:8000." This keeps Postman environments in sync without leaving your editor.
  • "Look at my Django Customer model and serializer, infer the OpenAPI definition, and create a Postman API spec and collection called "Customer API." This scaffolds a ready-to-test Postman collection directly from your backend code.
  • "Update the Customer API Collection documentation to reflect the latest endpoints and add request/response examples." This regenerates Postman API docs from current code without manual editing.
  • "Read the Customer API spec changes in Postman and reflect them in my Django Customer model and serializer." This propagates API contract changes back into backend code, keeping spec and implementation in sync.

RESOURCES

Go deeper on AI with Postman

Postmanaut working with server. Illustration.

Postman MCP Server

Give your AI coding agent secure access to your Postman API workflows.

See best practices →

Postman Agent Mode on desktop. Illustration.

Agent Mode

Have Postman build and test your APIs for you, with AI in the loop.

Learn more →

Postman handyman illustration.

MCP Generator

Turn a public API in the Postman API Network into an MCP server.

Learn more →

FAQ

What is MCP?

Model Context Protocol (MCP) is an open standard that enables applications to share context with large language models (LLMs), making it easier to build smart agents and automated workflows.

Which coding agents does Postman's remote MCP server work with?

You can use Postman's MCP server with all coding agents that support MCP.

How do the Postman API and the AI agent communicate?

Secure communication happens over HTTPS without extra setup required on the agent side. This is ideal for IDE integrations and cloud-hosted agents. There is also a local option via stdio.

What tool configurations are supported?

Postman supports the following tool configurations:

  • Minimal — (Default) Only includes essential tools for core Postman operations. This offers minimal token consumption and simplifies use for those who only need core Postman operations. Ideal for users who want to run collections or modify a single Postman element, such as collections, workspaces, or environments.
  • Code — Includes tools to generate idiomatic, well-organized client code from public and internal API definitions. This configuration is ideal for users who need to consume APIs or simply get context about APIs to their agents.
  • Full — Includes all available Postman API tools (100+ tools). Best for users who want to automate more advanced workflows.

What are some prompts I can use with Postman's MCP server?

Try these example prompts to get started:

  • "build a demo of accepting a payment using the stripe api. use postman"
  • "what statuses can the google maps geolocation api return? use postman"
  • "write a script that lists all our notion pages. use postman"
  • "create a client for our internal acme sync api. use postman"

The "use postman" phrase is optional but ensures your coding agent uses the Postman MCP Server tools. To make this automatic, add a rule to your coding agent:Always use the provided Postman MCP tools when I need to explore APIs, answer questions about APIs, or generate client code to consume APIs.

How is Postman protecting my data?

Your data is safe with Postman. We protect it carefully, just as promised in our Terms of Service and Privacy Policy. Visit our Security page to learn more.

How much is my monthly usage? Are there rate limits?

MCP tool calling counts towards your plan's monthly limits for calls to the Postman API.

Can I use Postman to generate MCP servers for my own internal APIs?

You can use Postman Agent Mode (currently in beta) to generate MCP servers for your internal APIs. Watch this demo to learn how.

Can I use Postman to generate MCP servers for public APIs?

You can use Postman's MCP Generator to choose public API requests from the Postman API Network and generate MCP servers for them. For each request, Postman creates a tool. Learn more.

How can I use Postman to test MCP servers?

You can use Postman to create MCP requests to experiment, test, and evaluate different MCP servers. Learn more.

How can I use Postman to discover MCP servers?

Browse this curated list of MCP servers to use within Postman.

How do I share feedback with the Postman team?

We value your feedback! Please share it with us by creating a GitHub issue.

Learn more about the Postman MCP Server

Chat with a Postman expert to learn how the Postman MCP Server can give your AI agents trusted, secure access to your API workflows—collections, environments, workspaces, and more.

Have feedback for us? Create a GitHub issue.

Postman MCP Server platform illustration