Run a pipeline manually
This guide shows you how to run a pipeline manually.
Start a run
Section titled “Start a run”- Navigate to Pipelines
- Find the template you want to run
- Click Run (or open the template and click Run pipeline)
- If the template has inputs, a form appears — fill them in
- Click Start run
The pipeline engine creates a PipelineRun record and begins executing steps.
Monitor the run
Section titled “Monitor the run”The run detail view shows:
- Overall status: running, completed, failed
- Step progress: each step’s status, input, and output
- Timeline: when each step started and finished
- Cost: cumulative token cost across all agent steps
Step statuses
Section titled “Step statuses”| Status | Meaning |
|---|---|
| Pending | Waiting for upstream steps |
| Running | Currently executing |
| Completed | Finished successfully |
| Failed | Ended with an error |
| Skipped | Skipped by a router or skip-propagation |
Cost tracking
Section titled “Cost tracking”Agent steps accumulate token costs. The run’s totalCostCents and totalTokens are updated as steps complete.
Inspect step output
Section titled “Inspect step output”Click any step to see:
- Input JSON: what was passed to the step
- Output JSON: what the step produced
- Error message: if the step failed
- Duration: wall-clock time
- Attempt count: how many times the step was tried (if retry policy is set)
Handle failures
Section titled “Handle failures”If a step fails:
- Check the error message in the step detail
- If the step has a retry policy, it may retry automatically
- If
continue-on-failis set, downstream steps proceed despite the failure - If neither, the run fails
Failed runs can be inspected but not resumed from the UI. Fix the template and start a new run.
Dry run
Section titled “Dry run”Enable Dry run when starting a run to see what would happen without actually executing steps. Dry runs:
- Evaluate the DAG structure
- Render template placeholders
- Show which steps would run and with what inputs
- Don’t dispatch agents or run scripts