ci(repo): fix codex-action v1.12 hang (downgrade to v1.11), adjudicate on >=1 review - #6380
Merged
Conversation
Contributor
Supabase CLI previewnpx --yes https://pkg.pr.new/supabase/cli/supabase@cc8ae01e22be790f78b5bc9d817f972c521f7435Preview package for commit |
jgoux
approved these changes
Aug 28, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Follow-up to the AI-review pipeline. Two changes.
1. Downgrade
codex-actionv1.12 → v1.11On the large PR (#6366, ~130k-token diff) the
codex-reviewstep ran to completion — Codex finished the turn, wrote its output file, printed its final message and token count — then sat idle until the 45-minute job timeout, discarding a completed review. That is a confirmed v1.12 regression: openai/codex-action#150 ("v1.12: Linux run never returns after the turn completes; job dies on timeout with the output file already written"). The reporter confirms v1.11 handles the same heavy workload cleanly, and there is no released fix above v1.12.v1.11 (
52fe01ec…) supports every input we use (safety-strategy,sandbox,output-schema-file,output-file,codex-version,working-directory,effort), so this is a drop-in pin change in both Codex jobs.2. Adjudicate on ≥1 independent review (graceful degradation)
Previously
adjudicaterequired BOTHclaude-reviewandcodex-reviewto succeed, so one flaky model job sank the whole review. Now it runs when at least one independent pass succeeded: each findings download is guarded by its job's result, and the stage step substitutes an empty findings set for any review that didn't complete, so the adjudicator reconciles one or two. The prompt notes the one-review case and records it in its summary.Together: a Codex hiccup no longer wastes a 45-minute run or blocks Claude's (working) review from being posted.