Skip to content

review-pr: harden Pre-Verdict Audit against comment-quality halo effect - #63

Draft
warp-agent-staging[bot] wants to merge 1 commit into
eval-base/019ff1e1from
factory/review-pr-audit-comment-halo-019ff1e1
Draft

review-pr: harden Pre-Verdict Audit against comment-quality halo effect#63
warp-agent-staging[bot] wants to merge 1 commit into
eval-base/019ff1e1from
factory/review-pr-audit-comment-halo-019ff1e1

Conversation

@warp-agent-staging

Copy link
Copy Markdown
Contributor

What

Hardens the ## Pre-Verdict Audit Comments bullet in .agents/skills/review-pr/SKILL.md.

Why

A reviewer using this skill let a comment's writing quality and technical accuracy ("well-written, explains a genuinely hard bug") stand in for actually checking it against the target repo's commenting guidelines, and missed confirmed rule violations as a result.

The revised bullet:

  • Requires enumerating every added/changed comment individually with its file:line, so none get skimmed past as part of a holistic impression.
  • Adds a fallback for repositories with no written commenting guidelines: judge against the commenting distribution of existing code (density, tone, what existing comments explain vs. omit).
  • Explicitly forbids treating writing quality, technical accuracy, or the importance of the described issue as a mitigating factor for a guideline violation.

Deliberately generic: no specific rule names or categories are introduced, so the skill stays portable across repos with or without explicit commenting guidelines.

Scope

Documentation only, one line, one section. The schema, severity labels, safety rules, evidence rules, suggestion-block constraints, and diff-line-annotation contract are untouched.

Require enumerating each added/changed comment individually with its
file:line, and forbid treating a comment's writing quality, technical
accuracy, or the importance of the issue it describes as a mitigating
factor for a guideline violation. Adds a fallback for repositories with
no written commenting guidelines: judge against the commenting
distribution of existing code.

Co-Authored-By: Warp Agent <agent@warp.dev>
@warp-agent-staging warp-agent-staging Bot added the factory-ab-eval Factory A/B evaluation replay label Aug 13, 2026
@warp-agent-staging

Copy link
Copy Markdown
Contributor Author

This PR was generated with Warp.

View run View conversation

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

factory-ab-eval Factory A/B evaluation replay

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant