Put it to work
Start agents through an API trigger
Make a published agent part of your own application or service.
Call a published agent from your own product or backend.
What this changes for your team.
Machine callers authenticate with a personal API key and POST a typed request to an API trigger. Authorization is rechecked at fire time, so holding a key never widens the caller beyond its owner's trigger access. Every key expires - after 90 days unless you choose otherwise, and at most a year - and opens only the scopes you ticked when you created it, so a key made to fire triggers cannot build agents. Rotating a key replaces its secret and keeps its name and scopes.
How it works in practice.
- 01
Publish the agent and subscribe it to an API trigger.
- 02
Create a key with the scopes and expiry the system needs, and store it only in your backend.
- 03
POST the input, retain the returned run identifier, and inspect or receive completion through the API surface.
What you can plan around.
What to expect, including limits that matter to your team.
Personal trigger keys act with their owner's permissions, rechecked on each fire.
A key carries an expiry and a scope list; a request outside its scopes, or after its expiry, is answered as if the route did not exist.
Authorization-shaped misses return a uniform not-found response rather than revealing inaccessible resources.
Calls refused before they reached any trigger are counted by reason on the API keys page, so a system still pointed at a retired key or deleted trigger is visible.
Bring one real process
See how Yekar.AI fits the way you work.
Start with a job your team already owns, plus the tools and decisions around it.
Get started