New Postman AI Features for Quality Engineers

I’ve done a few posts recently which document my continuing explorations into the world of AI for quality engineering tasks.

This will be a very short post, but there are a couple of new developments in Postman that I’ve been meaning to take a look at for the past few weeks. They recently introduced some tools to help support AI Agent connectivity and creation.

The two things I had a look at were:-

Postman Tool Generation API – using this in-built tool and a few drop downs, auto-generates you some boiler plate code you can use to integrate any of the 100K+ APIs into an AI Agent or LLM. Early days, but could be a real time saver if you wanted to try out any public APIs e.g. Mailinator. Only current code selections are JavaScript and Typescript but sure this will expand in time.

Postman AI Protocol – instead of creating a new request, workspace or collection, you now have the option of selecting “AI”. This allows you to create a single prompt that you can tweak and reuse across LLMs just by changing the model. See the video below where I try to use Anthropic creds for an OpenAI request, then without tweaking anything but the model name send the correct request.

There is also a Flow which provides outputs when several models are sent the same information – really handy if you’re testing model outputs.

Happy testing!