code-review

Automate adversarial code review with parallel analysis layers and structured triage.

3|Updated Jun 23, 2026
One-click install
npx skills add https://github.com/WritzProtocol/writz --skill code-review-writzprotocol
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/WritzProtocol/writz/tree/main/.agents/skills/code-review
Command: npx skills add https://github.com/WritzProtocol/writz --skill code-review-writzprotocol

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a comprehensive code review process that uses parallel review layers and structured triage to ensure code changes are thoroughly analyzed and actionable insights are provided.

Core Features & Use Cases

  • Adversarial Review Layers: Utilizes parallel review layers (Blind Hunter, Edge Case Hunter, Acceptance Auditor) to analyze code changes from different perspectives.
  • Structured Triage: Categorizes findings into actionable categories like 'decision_needed', 'patch', 'defer', and 'dismiss'.
  • Use Case: Ideal for software development teams to ensure code quality and maintain codebase integrity during the development process.

Quick Start

Run 'code-review' to initiate a code review session.

Frequently Asked Questions about code-review

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

FAQPage Schema
What is adversarial code review and how does it analyze code changes?

Adversarial code review analyzes code changes using parallel review layers like Blind Hunter, Edge Case Hunter, and Acceptance Auditor. These layers examine modifications from different perspectives to identify software quality issues and provide actionable insights.

How do I automate code analysis for software quality assurance?

You can automate code analysis by running the 'code-review' command to initiate a session. The process uses Python scripts and configuration files to execute structured triage across multiple review layers for comprehensive code quality assurance.

Do I need Python to run the adversarial code review process?

Yes, Python is required to run the adversarial code review process. The Skill relies on Python scripts and configuration files to execute its parallel review layers and structured triage for analyzing code changes.

How are code review findings categorized during structured triage?

During structured triage, code review findings are categorized into actionable groups: 'decision_needed', 'patch', 'defer', and 'dismiss'. This classification helps software development teams prioritize and address code quality issues effectively.

What is the best way to review code changes for edge cases and acceptance criteria?

Using parallel adversarial review layers like Edge Case Hunter and Acceptance Auditor is an effective way to review code changes. They analyze modifications from different perspectives to ensure thorough software quality analysis and codebase integrity.

Can I use this code review process for my software development workflow?

Yes, this adversarial code review process is ideal for software development teams. It ensures code quality and maintains codebase integrity during development by providing structured triage and actionable insights from multiple review perspectives.