skills.rest
looooonk
looooonkCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

code-review

Orchestrates subagents to review pull requests and report numbered findings.

5|1|Updated Jun 26, 2026
Search Tags:#code review#markdown#pull request#github#subagents#orchestration#findings
One-click install
npx skills add https://github.com/looooonk/better-codex --skill code-review-looooonk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/looooonk/better-codex/tree/main/.codex/skills/code-review
Command: npx skills add https://github.com/looooonk/better-codex --skill code-review-looooonk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Reviewing a pull request thoroughly requires checking many dimensions at once, and a single-pass review often misses issues. This Skill coordinates multiple specialized code-review subagents so every concern is covered and every finding is reported. ## Core Features & Use Cases - Parallel Subagent Reviews: Spawns one subagent per code-review-* skill, each with high reasoning effort, to cover distinct review dimensions. - Complete Findings Report: Aggregates every issue from every subagent into a numbered Markdown list, each with a specific file path and line number. - GitHub Integration: Adds a code-reviewed label when the reviewer is the pull request owner, and never posts comments unless explicitly asked. - Use Case: Before merging a large pull request, run this Skill to get a consolidated, numbered list of all issues found across security, style, and logic review passes. ## Quick Start Run a final code review on this pull request and report every finding with file paths and line numbers.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I run a comprehensive code review on a pull request?▼

Invoke the code-review skill on the pull request. It spawns one subagent per specialized code-review skill, then aggregates every finding into a numbered Markdown report with file paths and line numbers.

How does the code review orchestrator use subagents?▼

It launches one subagent for each code-review-* skill, passing the full skill path and using high reasoning effort. Each subagent reviews independently, and the orchestrator returns all findings without filtering.

Does the code review post comments on GitHub pull requests?▼

No, it does not leave GitHub comments unless explicitly asked. It only adds a code-reviewed label when the user running the review is the pull request owner.

What format do code review findings use?▼

Findings are reported in raw Markdown, numbered for easy reference. Each finding must include a specific file path and line number so issues can be located directly.

When should I not use the code-review orchestrator?▼

Avoid it for quick single-concern checks, since it runs every code-review skill via subagents and is designed for final comprehensive reviews. Use an individual code-review skill for targeted feedback.

Related Skills

Explore More Software Engineering Skills→
MS33834MS33834

Single-Flow PR Yeet

Stages, commits, pushes, and opens a GitHub pull request for a linear change.

Community
Advanced
FastLEDFastLED
7.5k

address-reviews

Fetch, classify, fix, and reply to CodeRabbit review comments on GitHub pull requests.

Official
Intermediate
uzumaki-incuzumaki-inc

health

Run code health checks and compute a weighted 0-10 score.

Official
Advanced
FerroxLabsFerroxLabs
21

ferrox-ns-review

Route review and quality intents to the correct specialized Ferrox skill.

Community
Basic