About Postman AI features

Postman’s AI features enable you to perform tasks faster, automate your workflows, and build your own AI models.

Agent Mode

With Agent Mode, you can turn your words into action across the API lifecycle. Send requests, fix errors, update tests, and more, using natural language.

The Postman CLI and coding agents

After installing the Postman CLI, you use the postman init command to set up an AI-ready API project in your local repository. In a single step, it creates the Postman Native Git project structure, adopts an existing API specification if your repository has one, and installs agent skills under postman/skills/, with one SKILL.md per skill and a root AGENTS.md that points agents to them.

For more information, see postman init.

AI models and Model Context Protocol (MCP) servers

You can experiment, test, and evaluate AI models and Model Context Protocol (MCP) servers. Use Postman to add an AI model of your choice to your project, such as one from OpenAI, Anthropic, or Google. Or, add an MCP server from the community—or build your own.

Start by sending requests to AI models. You can compare models based on their responses, response times, and token counts. Then, send requests to MCP servers. You can compare servers based on their tools, resources, and prompts. You can also combine the two and see how an AI model uses an MCP server to enhance a response.

With Postman’s MCP Generator, you can create your own MCP server with public APIs from the Postman API Network, and use Postman to improve your server’s developer experience.

Postman Flows AI

If you prefer a visual, low-code editor, you can use Postman Flows to experiment, test, and evaluate AI models and MCP servers. Or use Flows to build your own MCP server.

Postman AI demos and examples

See Postman’s AI tools in action in the Postman AI Tool Builder public workspace. Then, explore the following guided demos and start building your next AI agent.