Work with Postman from your agent

After you install the Postman plugin, describe what you want in natural language. Your coding agent can sync your project with a Postman Workspace, discover public APIs, create mock servers, or set up monitors.

You don’t need a Postman account for local work. Tasks that use the cloud, such as syncing a workspace or pushing a mock, prompt you to sign in to Postman first.

The following examples show tasks you can describe to your agent.

Build and test APIs

Your agent builds and runs API collections with the Postman CLI, saving each one as Native Git v3 files you can review and commit alongside your code.

Simulate your API’s dependencies

Your agent finds the services your code calls, gives each one a mock, and runs your real service against all of them from one local simulation. It reads the simulation’s log to show what each dependency returned, and can fail any of them with latency or errors to test how your code copes.

Investigate API dependencies

Questions about what depends on an API, who owns it, or what a change could break are answered with Postman Context Graph, a dependency map of your organization’s API ecosystem. Your agent queries it with the postman context-graph commands, so prompts like the following work without you pointing the agent at each service yourself.

Integrate public APIs

Requests to find and use a public API are handled with Orbit, which searches publicly available APIs for tasks like weather, payments, and messaging. For the API you choose, Orbit gives your agent what it needs to integrate the API, so you don’t have to find and read the provider’s documentation yourself.