Postman Monitors for coding agents and developers

Synthetic API monitoring runs scripted API requests and assertions against your endpoints on a schedule, from outside your system, so you find out about failures and slowdowns before your users do. Postman Monitors is a synthetic API monitoring product: a monitor runs a collection or API specification, including your test scripts and chained requests, and reports pass or fail for every run. It checks APIs, not browser sessions.

An uptime check tells you whether one URL answered. A Postman monitor runs your real requests and assertions in order, so it can tell you whether a user can sign in and then read their data. If step one passes and step two breaks, a single-endpoint check misses it and a monitor catches it.

Yes. With the Postman plugin and CLI, agents can create, update, run, pause, resume, and delete monitors, list run history, and read results as JSON using postman monitor commands. Your agent shows you the schedule, regions, and estimated usage before it creates a monitor.

You can try monitors in your browser with no account using the free tools. The postman init command starts a guest session, and the postman monitor commands accept it. Sign up when you want to keep your monitors and share them with your team.

Use Private API Monitoring. Start a runner with the Postman CLI inside your VPC or private network, then run monitors from it. Your endpoints aren't exposed publicly. For APIs behind a firewall that you can allowlist, you can instead run monitors from static IPs.

Yes. Pick one or more Postman Cloud regions, or a runner in your own network, when you create a monitor. Each request in each region counts as one monitoring call toward your usage. Run postman runner regions to see what's available.

Add email recipients (up to five from the CLI), or connect Slack, Microsoft Teams, PagerDuty, or a webhook. You can retry failed runs up to twice and mute repeat notifications after a set number of consecutive failures.

Yes. Trigger a monitor with postman monitor run after a deploy and use the result to pass or fail the build, or let the schedule keep watching after the release.

Each team can have up to 300 active and paused monitors. On a free plan, monitors pause for the rest of the billing period when you reach your monthly monitoring limit. On paid plans, overages are on by default and billed pay-as-you-go, and you can turn them off or buy blocks of calls in advance.