Run Test
Before relying on a schedule to run unattended, use the Run Test button on the schedule detail page to trigger an immediate execution.
A test run works exactly like a real scheduled run — same agent, same prompt, same project, tools, and datastores — so the output you see is what you will get on the actual schedule. The only differences:
Test runs do not count toward your monthly run limit.
Test runs are marked with a Test badge in the run history so you can tell them apart.
Test runs don't change when the next scheduled run happens, and don't affect the schedule's failure count.
If your schedule is paused or completed, you can still run a test without resuming it. It stays paused or completed afterwards.
After clicking Run Test, a confirmation dialog appears. Once confirmed, the run enters the Queued state, then moves to Running. While running, you can open the live chat to watch the agent work in real time. You can also cancel a test run while it is queued or running.
Tip: Always run at least one test before relying on a new schedule. This lets you verify the prompt, check the output quality, and catch issues before a real run.
Run History
Every schedule has a run history section showing all past executions — both regular and test runs.
Each run shows:
Status — Success, Failed, Cancelled, Running, or Queued
Test badge — blue "Test" label if it was a test run
Timestamp — when the run started
Duration — how long the run took to complete
Tokens — input and output token usage
View chat — link to open the conversation the run created
Understanding Run Statuses
Status | Meaning |
Queued | The run has been dispatched and is waiting to start execution. |
Running | The agent is actively processing the prompt. |
Success | The run completed without errors. The conversation is available to view. |
Failed | The run encountered an error. Check the error message in the run details. |
Cancelled | The run was manually cancelled before completion (test runs only). |
Understanding Schedule Statuses
Status | Meaning |
Active | The schedule will run at its next due time. |
Queued / Running | A run of this schedule is in progress. |
Paused | Paused by you or automatically after failures. No runs happen until it's resumed. |
Completed | The schedule passed its end date and has no more runs. Edit it and extend or remove the end date to start it again. |
Notifications
If you enabled notifications when creating the schedule, you will be notified each time a run finishes:
Email — a summary email is sent to all listed recipients. It includes the schedule name, run status, the agent's output text, and a link to view the full conversation. Emails are sent from [email protected].
Notifications are sent for both successful and failed runs, so you always know what happened.
Usage Limits
Limit | Default | Resets |
Active schedules per member, per workspace | 20 | — |
Successful scheduled runs per member, per workspace, per month | 100 | Start of each calendar month (UTC) |
Active schedules — the maximum number of schedules you can have running in a workspace at one time. Paused and completed schedules don't count toward this limit, so you can pause a schedule to free a slot. Resuming a schedule takes a slot again.
Monthly runs — each successful (non-test) scheduled run counts against this limit. Failed runs and test runs don't count. When the limit is reached, further scheduled runs for you in that workspace are skipped until next month — the schedule stays Active, no run is recorded, and it picks up again automatically once the limit resets.