Agent Ci Dev

---
description: Run local CI to verify changes before completing work
aliases: [validate]
---

// turbo-all

1. Run agent-ci in the **background** with the NDJSON event stream enabled so you can monitor and react to failures:

```bash
pnpm agent-ci-dev run --all -q -p --json 2>&1

Use run_in_background: true on the Bash tool. This returns an output file path. The launcher detaches automatically when stdout isn't a TTY; the foreground process exits 77 the instant a step pauses, while the worker keeps the container alive for retry.

2. Set up a Monitor on the output file to catch pause/finish events from the NDJSON stream:

tail -f <output-file> 2>/dev/null | grep --line-buffered -E '"event":"(run\.paused|run\.finish)"'

run.paused carries the runner name and retry_cmd you'll need in step 4. run.finish carries status: passed|failed.

3. Wait for either a monitor event (pause/finish) or a background task completion notification. If run.finish reports status: passed (or the background task exits 0 with no pause), stop the monitor with TaskStop and you're done.

4. If a step fails, the runner pauses and waits (foreground exit code 77; run.paused event in the log). CI was passing before your work started, so the failure is caused by your changes. Investigate and fix it:

• Read the output file for the full failure details.
• Identify and fix the issue in your code.
• Retry the failed runner (run in background and monitor the output):

pnpm agent-ci-dev retry --name <runner-name>

• If the fix requires re-running from an earlier step:

pnpm agent-ci-dev retry --name <runner-name> --from-step <N>

• Monitor the original output file for the retry results.
• Repeat until the job passes.

5. Once all jobs have passed, you're done.