quality-review-feedback

Classify code review feedback as accept, clarify, or reject with technical reasons.

1|Updated Mar 18, 2026
One-click install
npx skills add https://github.com/xxih/ai-harness-zh --skill quality-review-feedback
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: quality-review-feedback
Source: https://github.com/xxih/ai-harness-zh/tree/main/packages/quality-workflows/targets/codex/skills/quality-review-feedback
Command: npx skills add https://github.com/xxih/ai-harness-zh --skill quality-review-feedback

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ambiguous or external code review comments often lead to either blind acceptance or unnecessary changes; this Skill structures the handling of review feedback so decisions are technically justified and recorded. It helps avoid superficial agreement, prevents unverified edits, and ensures each comment is evaluated against the codebase and tests.

Core Features & Use Cases

  • Structured Reading & Paraphrase: Read the full feedback and restate each technical point in your own words to ensure understanding.
  • Verification & Evaluation: Cross-check suggestions against the repository, tests, and contextual constraints to decide whether a suggestion is correct, needs clarification, or should be rejected.
  • Actionable Outputs & Recording: Produce a per-item decision list (accept/clarify/reject) with technical reasons, implement fixes one by one, and write a summary to the current task record or .quality/quality-check.md.
  • Use Case: When receiving an external review with multiple unclear items, use this Skill to validate each suggestion, avoid regressions, and produce a documented response and change set.

Quick Start

Please review the provided code review comments, identify which items to accept, clarify, or reject with technical reasons, and write a concise Review Feedback summary into .quality/quality-check.md.

Frequently Asked Questions about quality-review-feedback

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I evaluate ambiguous code review feedback against my repository?

To evaluate code review feedback, read the full comments, paraphrase each technical point, cross-reference suggestions against your repository artifacts and tests, and classify each item as accept, clarify, or reject based on technical impact.

What is the best way to structure external code review comments into actionable decisions?

Structuring code review comments involves processing incoming feedback, verifying each suggestion against the codebase, and producing a per-item decision list with technical reasons and implementable fixes recorded in a quality-check file.

Can I use this skill to avoid blind acceptance of unclear reviewer suggestions?

Yes, this skill helps avoid blind acceptance and unverified edits by performing structured reading, evaluating the impact of reviewer suggestions on existing behavior, and justifying each accept, clarify, or reject decision technically.

How do I generate quality records from code review verification?

You generate quality records by implementing verified fixes one by one after evaluating review feedback, then writing a concise Review Feedback summary of the decisions into the current task record or .quality/quality-check.md file.

Does this skill work for code review scenarios receiving external and unclear comments?

Yes, this skill is designed specifically for code review scenarios in software repositories receiving external or ambiguous reviewer comments that require verification against code, tests, and project context before making changes.

When should I not blindly accept code review feedback without verification?

You should not blindly accept feedback when comments are ambiguous or external, as superficial agreement can lead to unnecessary changes or regressions; instead, cross-check suggestions against the repository and tests to decide whether to accept, clarify, or reject.