review

Analyze pull request diffs for code quality and security issues.

7|Updated Oct 20, 2025
One-click install
npx skills add https://github.com/borkweb/skills --skill review-borkweb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/borkweb/skills/tree/main/skills/gstack/review
Command: npx skills add https://github.com/borkweb/skills --skill review-borkweb

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires git, grep, bash, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill ensures code quality and security by conducting a thorough pre-landing PR review, identifying potential issues that tests might miss, and providing a structured approach to code analysis and review.

Core Features & Use Cases

  • Pre-Landing Code Review: Analyzes diffs against the base branch for SQL, migration, auth/permission, error handling, API contract, and LLM trust boundary issues.
  • Design Review: Evaluates frontend changes against design guidelines and best practices.
  • Adversarial Review: Identifies potential production failures by thinking like an attacker.
  • Fix-First Review: Applies fixes to mechanical issues and prompts user for input on more complex ones.
  • Documentation Staleness Check: Ensures documentation reflects the latest code changes.
  • Post-Fix Verification: Verifies fixes and tests to ensure no regressions occurred.
  • Use Case: When reviewing a PR for code changes, run the review skill to analyze the code for issues and provide a detailed review report.

Quick Start

Run the '/review' command when asked to review a PR or before merging code changes.

Frequently Asked Questions about review

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

FAQPage Schema
How do I perform a pre-landing code review for security and quality issues?

A pre-landing code review analyzes git diffs against the base branch to identify SQL, migration, auth, and API contract issues. It evaluates structural problems, checks frontend design guidelines, and performs adversarial review to catch potential production failures before merging.

What is adversarial code review and how does it find production failures?

Adversarial code review identifies potential production failures by analyzing diffs like an attacker. It targets LLM trust boundaries, permission flaws, and error handling weaknesses that standard tests might miss, providing a structured security audit before code lands.

How do I check if my documentation is stale after updating code?

To check if documentation is stale, a code review checks the diff to ensure documentation reflects the latest code changes. It identifies outdated references and applies mechanical fixes directly, prompting for user input on more complex documentation updates.

Do I need git and bash to run an automated PR review?

Yes, automated PR review requires git, grep, and bash to analyze diffs and execute structural checks. You may also need additional tools installed in your environment to perform specific frontend design or security audits.

Can a PR review automatically fix mechanical code issues before merging?

Yes, a fix-first PR review applies fixes to mechanical code issues automatically. For complex structural or security problems, it generates a detailed review report and prompts you for input before applying any changes to prevent regressions.

What's the best way to verify code fixes and prevent regressions after a review?

The best way to verify fixes and prevent regressions is post-fix verification. After applying changes from the PR review, it runs tests to ensure no regressions occurred and validates that the original security and quality issues are fully resolved.