review

Audit pull requests for correctness, UX, integration contracts, and KPI impact.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/sujithatzackriya/skills --skill review-sujithatzackriya
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/sujithatzackriya/skills/tree/main/skills/review
Command: npx skills add https://github.com/sujithatzackriya/skills --skill review-sujithatzackriya

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you prevent merged code defects by producing a structured, end-to-end pull request review report that checks correctness, usability/UX, integration contracts, data accuracy, regression surface, and KPI impact.

Core Features & Use Cases

  • Deep PR audits: Uses an opt-in --deep mode that pins rigorous WORKFLOW-A frameworks and forces call-path tracing plus per-finding reverification.
  • Structured PR review reporting: Produces a markdown report with categorized findings, verification verdicts, and (in --deep) “Findings reverified” and “Frameworks used” in the TL;DR.
  • Scales to large PRs: Supports multi-agent parallel review for larger diffs and consolidates results.
  • Optional codex reconciliation: Runs or coordinates an independent codex review pass to compare findings and improve confidence.

Quick Start

Run: /review 123 and answer the codex opt-in prompt to receive a thorough PR review report.

Frequently Asked Questions about review

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

FAQPage Schema
How do I perform a deep audit on a pull request to check regression risks?▼

A deep pull request audit uses call-path sweeps and falsification tests to verify correctness, UX, integration contracts, and regression surface. You can trigger this rigorous review mode by using the optional --deep flag during your PR review.

What is the best way to review large pull requests with multiple agents?▼

Reviewing large pull requests is handled by multi-agent parallel review, which scans the diff concurrently and consolidates results. This scales the code review process to ensure comprehensive regression surface and data accuracy analysis.

Can I reconcile code review findings with an independent codex pass?▼

Yes, you can run an independent codex review pass to compare findings and improve confidence. This optional codex reconciliation is governed by an opt-in gate that prompts you before executing the comparative analysis.

Does the pull request review report cover UX verification and integration contracts?▼

The pull request review report explicitly audits UX verification and integration contracts. It produces a structured markdown document with categorized findings and verification verdicts covering these specific areas.

How do I start reviewing a pull request by branch or URL?▼

You can review a pull request by providing its number, URL, or branch name. The review process generates a structured markdown report auditing correctness, data accuracy, and KPI impact.

What frameworks are used during a deep code review?▼

A deep code review pins rigorous WORKFLOW-A frameworks to force call-path tracing and per-finding reverification. The specific frameworks used are listed in the TL;DR of the generated review report.