Reading a pipeline execution

Modified on: Tue, 7 Jul, 2026 at 7:48 PM

When a pipeline runs, its execution view shows you what's happening and what happened. This article covers how to read a run, follow its progress, and act on the result.


The execution view

A pipeline run looks a lot like a CI or GitHub workflow. A diagram shows the sequencing of steps, so you can see the shape of the run at a glance. For very large pipelines the diagram gets dense, but it still shows the order things happen in.


The pipeline advisor

As a pipeline runs, the pipeline advisor checks in periodically. It tells you how the run is going — what the agents are experiencing — and flags whether the pipeline itself could be more efficient. Think of it as a running read on both the work and the workflow.


Reports

A pipeline can output a report — a summary it writes to its output that shows up in the execution view. A code-review pipeline, for example, produces a review report.


Drilling into steps and transcripts

You can drill into any step to see how it went — its inputs, outputs, and full transcripts — as well as effective costs. Drill down as much or as little as you need to understand what a run did, or where it got stuck. Every run also lands on the Audit page, where everything that runs in the app is recorded for later review.


Following up on a result

From a finished run you have several options, depending on how the pipeline is set up:

  • Chat off the result — start a chat from the output to keep going.
  • Look at the report, or send the report or plan into a worktree chat.
  • Start over, or retry if the run failed.
  • Chain a follow-up pipeline from the run.

If you set a watcher on the run, it pings a chat when the pipeline finishes or fails, so you don't have to sit and watch it.


To dig into a run that went wrong, you can also ask the Pipeline Assistant (now available) or App Advisor — describe what happened and get help fixing the pipeline.


Next steps

  • Pipelines: overview — run modes and where pipelines live.
  • Anatomy of an implementation pipeline — what the stages of a deep run are doing.
  • Watchers — get pinged in-chat when a run finishes or fails.

Did you find it helpful? Yes No

Send feedback
Sorry we couldn't be helpful. Help us improve this article with your feedback.

Still need help?

Submit a Support Ticket

Join us on social media

with us