Run now and read the history
Trigger an immediate run and follow the real result of each execution in the history.
Run now
The Run Now button, in the routine’s detail, queues an immediate execution.
The return is immediate and is not the result. It confirms that the run entered the queue — the status that appears at that instant is the initial one, not the final one. The work is still going to happen.
To know the outcome, the path is the history: follow the execution and read the status and the error of the first item.
When the button is unavailable, it says why — a paused routine or an operation in progress, for example.
Execution history
The Run history section of the detail, and the executions tab of the screen, show each run:
| Column | What it says |
|---|---|
| Run | The run’s identification |
| Workspace | Where that execution happened |
| Spend | The associated cost |
| Tokens | The token consumption |
| Status | The outcome: completed, failed, unknown |
Without a filter, the history is general. With the filter by automation, only its runs.
How to use this in practice
- After creating a routine, run it once now. Before waiting for the schedule, a manual execution shows whether the request is clear and whether the environment responds.
- Read the status, not the “queued”. The queueing message says the run started, not that it succeeded.
- Read the error when it fails. The history keeps the reason; without reading it, the routine looks like it just “does nothing”.
- Check the consumption. Accumulated tokens and cost are the way to notice a routine that is running more than it should.
- A skipped run is a result. When the precheck fails, the execution is recorded as skipped — that is the routine working as expected, not an error.
Scope and limits
- The initial status is not the outcome. Treating the immediate response as the result is the mistake this page exists to prevent.
- There may be pagination or an internal limit in the history. Do not treat the list as the complete record of every run to date.