review-thorough

Run the standard review and audit all bot-generated comments on GitHub pull requests.

Updated Apr 15, 2026
One-click install
npx skills add https://github.com/dgowrie/claude-workflows --skill review-thorough
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-thorough
Source: https://github.com/dgowrie/claude-workflows/tree/main/skills/review-thorough
Command: npx skills add https://github.com/dgowrie/claude-workflows --skill review-thorough

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Wrapper around the built-in /review command that additionally fetches and evaluates every bot-generated review on the PR — including threads marked resolved. Use when you want the standard /review flow plus assurance that prior bot feedback hasn't been silently dismissed.

Core Features & Use Cases

  • Run the built-in /review flow to capture the full summary, bugs, warnings, and suggestions.
  • Enumerate all bot-generated review threads (resolved or not) across Copilot, CodeRabbit, Codex, Sourcery, Gemini Code Assist, and other GitHub Apps, preserving thread and comment identifiers for traceability.
  • Evaluate each bot comment against the current diff, categorize as still valid, addressed, false positive, or style-only, and append a Bot review coverage section to the output for auditability.

Quick Start

Invoke the thorough PR review by triggering the skill with skill="review-thorough" and supply the PR number and repository context so it can run the full /review and audit all bot comments.

Frequently Asked Questions about review-thorough

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

FAQPage Schema
How do I audit resolved bot comments on a GitHub pull request?

To audit resolved bot comments on a GitHub pull request, enumerate all bot-generated threads including resolved ones, cross-reference them against the current diff, and categorize them as valid, addressed, or false positive for a comprehensive coverage report.

What is a comprehensive code review with bot feedback validation?

A comprehensive code review with bot feedback validation runs the standard review flow to capture bugs and warnings, then fetches and evaluates every bot-generated review comment against the current diff to ensure prior automated feedback hasn't been silently dismissed.

How do I check if CodeRabbit or Copilot review comments are still valid?

Checking if CodeRabbit or Copilot review comments are still valid requires evaluating each bot comment against the current pull request diff. This cross-referencing categorizes feedback as still valid, addressed, false positive, or style-only.

Does this bot review audit work with multiple GitHub Apps like Sourcery and Gemini Code Assist?

Yes, this bot review audit works with multiple GitHub Apps including Sourcery, Gemini Code Assist, Copilot, CodeRabbit, and Codex. It preserves thread and comment identifiers for all automated reviews across these platforms to ensure full traceability.

How do I get a standard PR review plus an audit of bot coverage?

To get a standard PR review plus an audit of bot coverage, run the built-in review flow to capture the full summary, bugs, and warnings, then append a Bot review coverage section that classifies all bot feedback against the current diff.