potiuk opened a new pull request, #63319:
URL: https://github.com/apache/airflow/pull/63319
## Summary
- Start LLM assessments in a background thread pool immediately after
deterministic checks complete, instead of blocking on them
- Process deterministically flagged PRs (CI failures, conflicts, unresolved
comments) interactively while LLM assessments run in the background
- Show LLM progress status between each deterministic PR prompt
(e.g., "LLM assessment: 3/10 done, 1 flagged, 7 in progress")
- After deterministic PRs are handled, show a Rich progress bar for any
remaining LLM assessments, then present LLM-flagged PRs for review
This overlaps human review time with LLM processing time, reducing the
total wall-clock time for a triage session.
## Test plan
- [ ] Run `breeze pr auto-triage` with mix of CI-failing and clean PRs
— verify deterministic PRs are presented immediately while LLM runs
- [ ] Verify LLM progress status appears between deterministic PR prompts
- [ ] Verify progress bar shows when waiting for remaining LLM assessments
- [ ] Verify LLM-flagged PRs are presented after deterministic PRs
- [ ] Run with `--check-mode=ci` — verify LLM is skipped as before
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---
##### Was generative AI tooling used to co-author this PR?
- [X] Yes — Claude Code (claude-opus-4-6)
Generated-by: Claude Code (claude-opus-4-6) following [the
guidelines](https://github.com/apache/airflow/blob/main/contributing-docs/05_pull_requests.rst#gen-ai-assisted-contributions)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
