code-review

Automate local code-quality review with dual sub-agents and human triage.

2|Updated Jun 15, 2026
One-click install
npx skills add https://github.com/jmerkow/my-agentic-stuff --skill code-review-jmerkow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/jmerkow/my-agentic-stuff/tree/main/skills/code-review
Command: npx skills add https://github.com/jmerkow/my-agentic-stuff --skill code-review-jmerkow

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill streamlines the process of local code-quality review, allowing developers to identify and address issues like style, slop, and structure without the need for a pull request.

Core Features & Use Cases

  • Dual Agent Review: Utilizes two independent sub-agents to identify issues, reducing the likelihood of oversight.
  • Human Triage: Empowers developers to make final decisions on findings, ensuring a balanced review process.
  • Use Case: Ideal for reviewing large codebases or complex projects where manual review is impractical or time-consuming.

Quick Start

Run the code-review skill to initiate a local code-quality review on your project.

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 local code review without creating a pull request?

To run a local code review without a pull request, you can use an automated skill that deploys dual independent sub-agents to identify style, slop, and structural issues directly on your local codebase.

What is dual-agent code review and how does it reduce oversight?

Dual-agent code review uses two independent sub-agents to analyze code quality simultaneously, cross-referencing findings to reduce oversight and ensure thorough issue identification before you proceed.

How do I triage code quality issues found by automated review agents?

You triage code quality issues found by automated review agents by reviewing the identified style and structural problems, then using human triage to make final decisions on which findings to address.

Can I use automated code review for large codebases where manual review is impractical?

Yes, you can use automated dual-agent code review for large codebases where manual review is impractical, as it efficiently identifies local code-quality issues without requiring a pull request.

What is the best way to check code structure and style locally before committing?

The best way to check code structure and style locally before committing is using an automated dual-agent review process that identifies issues and provides them for human triage.

Does local code review replace the need for a pull request?

Local code review does not replace a pull request but streamlines the process by allowing developers to identify and address style, slop, and structural issues locally before submitting code.