Watch a cloud run

A cloud run streams into a cloud pane, the same kind of pane a local agent runs in. You watch the output, send a follow-up, stop the run, and open the pull request from there. The pane's output is durable, so you can close it and open it again later and the run is still there.

While it prepares

Before any output appears, the pane shows what it is doing: checking cloud run readiness, then preparing the cloud workspace, then starting the agent. If a step runs long, the message switches to a "still" version so you know it has not stalled.

Once the agent starts, the pane shows "Waiting for terminal output" until the first output lands, then streams the run live.

Send a follow-up

While a run is live, you can message the agent from the pane. Write the follow-up and send it, and Coldtea saves it with the running cloud run. If the run cannot take input at that moment, the pane says so ("Cloud run is streaming. Follow-ups unavailable right now.") and your message waits as a queued message until it can be delivered.

Stop a run

Use Stop to end a live run. The button reads "Stopping" while it winds down. A stopped run keeps whatever output it captured up to that point.

When the run ends

A finished run opens a draft pull request. Open it to review the change. See review agent work for how to read one.

To keep going, use Continue. Coldtea starts a linked run that carries the transcript and the result context, so the next run builds on this one. A same-agent continuation keeps the parent run's model and reasoning effort; a different agent falls back to the CLI default. You can also start a fresh run from the task.

If a run does not start, the pane explains why: setup that is still missing, a failure message, or that it was stopped before any output arrived. Fix the setup and start again. See start a cloud run for the prerequisites.