chore: silence bot reviewers (Claude high-signal, CodeRabbit disabled) - #101
Merged
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
Bot reviewers were generating infinite nit-cycle reviews — each push triggered a fresh full-PR rescan that surfaced increasingly small style / wording / hypothetical-future-proofing findings, never converging on approval. This rewrites the Claude review workflow prompt to enforce a strict high-signal-only bar (correctness, security, contract, missing-test on mutating state, AGENTS.md hard rules including cross-platform parity) and require an explicit APPROVE / REQUEST_CHANGES / COMMENT outcome via `gh pr review`. Default is APPROVE — manufacturing nits to avoid approving is now explicitly called out as a bug. Added Bash(gh pr review:*) to the allowed tools. CodeRabbit and ChatGPT Codex auto-reviews are disabled at the integration level (CodeRabbit and ChatGPT settings respectively); no repo files needed. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
thomasluizon
force-pushed
the
chore/silence-bot-reviewers
branch
from
May 19, 2026 20:06
91e964a to
ab15ef4
Compare
|
Note Currently processing new changes in this PR. This may take a few minutes, please wait... ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
|
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
The chat-clarification PR (#100) generated 14 rounds of bot reviews and 101 resolved threads. Each push produced fresh findings — eventually all just style / wording / hypothetical-future-proofing. The loop never converges because bots are optimized to surface findings, not to approve.
This PR rewrites
.github/workflows/claude-review.ymlto enforce a strict high-signal-only bar (correctness, security, contract break, missing-test on mutating state, AGENTS.md hard rules including cross-platform parity) and requires Claude to pick one of three formal review states: APPROVE (default), REQUEST_CHANGES, or COMMENT. Manufacturing nits to avoid approving is explicitly called out as a bug.CodeRabbit and ChatGPT Codex auto-reviews are disabled at the integration level (their respective dashboards / app uninstalls), not via repo files.
Why now
Real correctness/security findings landed in rounds 1-6 of the chat-clarification review. Rounds 7-14 were all polish: focus-visible rings, transition-color flip-flops, Zod
.nullable().optional()modifier purity, magic-number-to-shared-constant migrations. None blocked merging.Test plan
Follow-ups (manual, not in this PR)
🤖 Generated with Claude Code