chore(ai-reviews): bump ai-workflows pin to v3.1.4 (silent-failure fix + clear outcome signaling)
Nobody has claimed this yet.
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 76/100
- Issue type
- Bug
- Clarity
- Mostly clear
- Activity status
- Active
- Tech stack
- github-actions
- Domain
- ci-cd
Research direction
Start with ai_claude-orchestrator.yml and ai_claude-backend-reviewer.yml, locating the referenced dotCMS/ai-workflows pins in both orchestrator jobs and the backend reviewer. Resolve the version discrepancy between the title's v3.1.4 and the body and changes' v3.1.3, then verify both workflow files use the intended release.
Written by the indexing model from the issue text.
Description
Summary
Bump the dotCMS/ai-workflows pin in the AI review workflows from @v3.1.2 to @v3.1.3.
Why
~43% of recent GPT-5.5 automatic reviews were posting "❌ Codex Review failed — job failed before producing output" and giving the PR no review. Root cause (fixed in ai-workflows#38, released as v3.1.3): on the Responses API, max_output_tokens caps the combined reasoning + answer tokens, so at reasoning_effort: medium GPT-5.5 sometimes spent the whole 2048 budget reasoning and returned status=incomplete with no visible text — which the executor treated as a hard failure.
v3.1.3 raises the default budget to 8000, captures the incomplete terminal event, retries once with a larger budget + lighter reasoning, and posts a clear diagnostic instead of a generic failure.
Validated e2e (before/after) on dotCMS/steve-quarterly-planning #105.
Changes
ai_claude-orchestrator.yml:@v3.1.2→@v3.1.3(both jobs)ai_claude-backend-reviewer.yml:@v3.1.2→@v3.1.3
- Dominant language
- Java
- Stars
- 970
- Forks
- 486
- Avg merge
- 3d 2h
- Merged PRs (30d)
- 164
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from dotCMS/core
-
dotCMS : Build Team : Falcon Type : Task
Difficulty 1/5 Under an hour Newbie friendliness 92/100
-
OKR : Customer Support Team : Maintenance
Difficulty 1/5 Under an hour Newbie friendliness 88/100
-
Team : Modernization
Difficulty 1/5 Under an hour Newbie friendliness 78/100
-
dotCMS : Workflow Team : Cloud Eng
Difficulty 2/5 1-3 hours Newbie friendliness 87/100
-
Team : Maintenance
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
infinispan/infinispan#18150 ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 84/100
-
untriaged
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
opensearch-project/k-NN#3597 ·
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 82/100