Webhook events
Every subscribable webhook event, grouped by the resource it reports on.
These are the event types an outbound webhook can subscribe to. New ones are added over time, so treat the list as open — capability discovery returns the current catalog for your workspace.
The console's test action sends ping, a delivery test
rather than a platform event. Your receiver should accept it and ignore it.
See Webhooks for delivery, signing, and retries.
Run progress
Follow agent and team runs from start to finish.
| Value | Meaning |
|---|---|
run.started | An agent or team begins working. |
run.completed | An agent or team completes its work. |
run.failed | A run ends before completing its work. |
run.cancelled | Someone or something stops a run. |
run.paused | A run stops temporarily and can continue later. |
run.resumed | Work restarts after a pause. |
run.requires_approval | A person must approve the next step before work continues. |
Problems during runs
Know when a run needs attention.
| Value | Meaning |
|---|---|
run.tool_call.failed | A tool returns an error while a run is working. |
run.policy_violation | A run attempts an action that a policy does not allow. |
run.resource.warning | A run is close to its configured resource allowance. |
run.resource.exceeded | A run uses all of a configured resource allowance. |
Pipelines
Follow automated pipelines through completion.
| Value | Meaning |
|---|---|
pipeline.completed | Every required pipeline step completes. |
pipeline.failed | A pipeline stops because one of its steps fails. |
Workforce runs
Follow coordinated workforce runs and conflicts.
| Value | Meaning |
|---|---|
workforce.run.started | A workforce begins coordinated work. |
workforce.run.completed | A workforce completes its coordinated work. |
workforce.run.failed | A workforce stops before completing its work. |
workforce.scope.conflict | Workforce members claim overlapping areas of work. |
Alerts
Forward alerts as soon as their conditions are met.
| Value | Meaning |
|---|---|
alert.triggered | A configured alert condition is met. |
Budgets and limits
Know before usage or spending stops important work.
| Value | Meaning |
|---|---|
budget.breach.warn | Spending reaches the budget's early warning threshold. |
budget.breach.critical | Spending reaches the budget's critical threshold. |
budget.hard_stop | A hard budget limit prevents more work from starting. |
run.limit_exceeded | A configured usage or concurrency limit prevents work. |