Monitor an API in Postman with AI
Schedule checks that verify what a caller actually needs, then read the evidence from the region, job and request that produced an alert.
Monitor an API in Postman with AI by deciding what would make a successful response useless to its caller. An order can return a success code and still omit the dispatch date. The api-monitoring skill helps an agent put that sort of assertion on a schedule and read the evidence from a particular run.
Location matters. A hosted runner may reach a public service, while an internal endpoint may need an authorised runner on the team’s network. Creating a monitor can also start the collection immediately, which is worth stating explicitly when the request was only to set up tomorrow’s check.
What it does
- Choose the cadence, time zone and intended alert recipients.
- Check that the selected collection is safe to run repeatedly in this environment.
- Compare regional results before treating an alert as a general outage.
- Follow a timed-out command to its existing job and terminal result.
- Update a schedule in place when the existing monitor history matters.
Install and use
- Provide an authorised collection, the official Postman tools and workspace access. For an internal service, use the registered team runner and protect its key.
- Ask, “Schedule a morning check of the test order endpoint, require a dispatch date and leave the first run until tomorrow.”
- Read the first completed job to establish where it ran and which assertions actually passed.
FAQ
Source
Made by Postman. Original skill: api-monitoring, published on GitHub under the Apache-2.0 licence.
View on GitHub