review

Generate structured code review reports for pull requests and local git branches.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/vitalwarley/hyprdots --skill review-vitalwarley
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/vitalwarley/hyprdots/tree/main/claude-global/skills/review
Command: npx skills add https://github.com/vitalwarley/hyprdots --skill review-vitalwarley

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manual code reviews are time-consuming, inconsistent, and often miss cross-PR patterns, convention drift, and reproducible verification steps; this Skill automates a disciplined, evidence-based review workflow and produces a persistent report artifact for asynchronous collaboration.

Core Features & Use Cases

  • End-to-end review orchestration: determines diff scope (unstaged, staged, branch, or PR), runs mechanical checks (linters, imports, tests), and performs verification passes when required.
  • Parallel analysis agents: launches specialized agents for code quality, silent failure hunting, type design, and comment quality to produce structured findings.
  • Report generation & provenance: writes a full review file to docs/reviews on the PR branch, includes spec compliance tables, root cause analysis, fix guides, and a short chat summary for posting to GitHub.
  • Use Cases: auditing a GitHub PR for spec compliance, verifying fixes on an R2 pass, or producing a review artifact for stacked PRs and cross-PR convention checks.

Quick Start

Ask the review skill to review PR 42 and save a structured report to docs/reviews for the branch.

Frequently Asked Questions about review

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

FAQPage Schema
How do I automate GitHub PR code reviews for spec compliance?

Automate GitHub PR code reviews for spec compliance by orchestrating mechanical checks, parallel analysis agents, and reproducible verification passes. The review process generates a structured markdown report artifact containing spec compliance tables and root cause analysis, saving it directly to the PR branch.

What is the best way to run linters and tests on a local git branch before creating a PR?

Run linters and tests on a local git branch by determining the diff scope and executing mechanical checks across the staged changes. The review workflow performs verification passes using reproducible commands to ensure code quality and convention adherence before PR creation.

How does parallel code review analysis find silent failures and type design issues?

Parallel code review analysis finds silent failures and type design issues by launching specialized agents that scan the diff for code quality, silent failure hunting, type design, and comment quality. These agents produce structured findings that feed into a comprehensive review report.

Can I generate a markdown review artifact for stacked PRs and asynchronous collaboration?

You can generate a markdown review artifact for stacked PRs and asynchronous collaboration by writing the full review file to docs/reviews on the PR branch. The artifact includes spec compliance tables, root cause analysis, fix guides, and a short chat summary for GitHub.

Does automated PR review work with unstaged changes or only fully staged branches?

Automated PR review works with both unstaged and fully staged changes by determining the diff scope across unstaged, staged, branch, or PR contexts. It applies consistent mechanical checks, linters, and tests across all specified files within the engineering codebase.

Why do I need reproducible verification commands during a code review?

Reproducible verification commands are needed during a code review to verify fixes on an R2 pass and ensure evidence-based validation. The review workflow runs mechanical checks and verification passes using these commands to prevent false positives and confirm spec compliance.