rune-review-intake

Classifies, verifies, and implements code review feedback with pushback and rejection tracking.

1|Updated Mar 22, 2026
One-click install
npx skills add https://github.com/dangvu008/VietTruyen --skill rune-review-intake-dangvu008
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rune-review-intake
Source: https://github.com/dangvu008/VietTruyen/tree/main/.agents/skills/rune-review-intake
Command: npx skills add https://github.com/dangvu008/VietTruyen --skill rune-review-intake-dangvu008

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When you receive PR comments or external code review feedback, blindly implementing every suggestion can break existing functionality or introduce hallucinated APIs. This Skill enforces a verification-first discipline: understand all feedback, verify claims against the actual codebase, then act with evidence-based verdicts. ## Core Features & Use Cases - Feedback Classification: Sorts every review item into BLOCKING, BUG, IMPROVEMENT, STYLE, or OPINION with P0–P4 priorities before any code changes. - Verification & Pushback: Checks reviewer claims against the real codebase using scout and grep, pushes back with technical evidence when suggestions are wrong, and applies YAGNI checks for unused features. - Rejection Knowledge Base: Writes durable .out-of-scope/<slug>.md records for rejected enhancements so decisions are never re-litigated in future sessions. - Use Case: A teammate leaves 8 comments on your pull request. The Skill classifies each one, verifies the claims, fixes the two real bugs, pushes back on one incorrect suggestion with test evidence, and records an out-of-scope feature request in the rejection KB. ## Quick Start Process the review comments on my current pull request and give me a verdict and action plan for each item.

Frequently Asked Questions about rune-review-intake

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

FAQPage Schema
How do I handle PR review comments before implementing changes?

Classify every comment by type (blocking, bug, improvement, style, opinion) and priority first, then verify each claim against the actual codebase before writing any code. Implement verified items in priority order and run tests after each individual fix.

How to respond to code review feedback professionally?

Start every response with an action verb such as Fixed, Reverted, Deferred, or Pushed back, followed by technical evidence. Avoid performative phrases like "great point" or "you're absolutely right" — signal with actions and reasoning, not pleasantries.

Should I trust AI-generated code review suggestions?

AI-generated reviews carry the lowest trust level and require double-checking of every suggestion due to high hallucination risk. Verify that referenced files, functions, packages, and APIs actually exist before implementing anything.

When should I push back on a code review suggestion?

Push back when a suggestion breaks existing functionality, ignores why the current implementation exists, proposes unused features (YAGNI), is technically incorrect for your stack, or conflicts with documented architectural decisions. Lead with test evidence and technical reasoning.

How do I track rejected feature requests from code reviews?

Write a durable record to an out-of-scope knowledge base file for every rejected enhancement, including the reasoning and conditions that would change the decision. Check for existing similar entries first and append repeat requests instead of creating duplicates.