code-review

Review code against SPEC.md and output approval.md or issues.md.

Updated Apr 28, 2026

Search Tags:#security #quality #review #issues #spec #code-review #approval

Ready to run

npx skills add https://github.com/PMKhai/claude-config --skill code-review-pmkhai

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/PMKhai/claude-config/tree/main/skills/code-review
Command: npx skills add https://github.com/PMKhai/claude-config --skill code-review-pmkhai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables automated, SPEC-driven evaluation of code quality by verifying correctness, security, and adherence to standards and then generating corresponding outputs.

Core Features & Use Cases

SPEC-driven code review against SPEC.md to ensure correctness, security, and standards.
Outputs either tasks/[task-id]/review/approval.md or tasks/[task-id]/review/issues.md based on findings.
Reads tasks/[task-id]/SPEC.md and code from tasks/[task-id]/code, enabling consistent, auditable reviews.

Quick Start

Run a SPEC-driven code review on the target repository and generate the appropriate approval.md or issues.md.

Frequently Asked Questions about code-review

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

FAQPage Schema

How do I automate code-quality reviews against a spec?▼

Automated code-quality reviews evaluate your repository code by reading the target SPEC.md file, verifying correctness, security, and standards, and then generating a formal review output.

What is SPEC-driven code review and how does it work?▼

SPEC-driven code review is an automated process that reads project specifications from SPEC.md and source code from the target repository to ensure code correctness and security before generating an audit report.

How do I generate an approval or issues report for my code repository?▼

To generate an approval or issues report, the automated review process evaluates your code against SPEC.md and outputs either an approval.md or issues.md file directly in the task review directory.

Can I use this automated review to check code security and standards?▼

Yes, automated SPEC-driven review checks code security and verifies adherence to project standards by reading SPEC.md and comparing it against the source code in your repository.

What directory structure is needed for automated SPEC.md code review?▼

The review process requires a specific directory structure with tasks/[task-id]/SPEC.md for specifications, tasks/[task-id]/code for source code, and tasks/[task-id]/review/ for outputting the generated approval or issues files.

When should I not use a SPEC-driven code review?▼

SPEC-driven code review should not be used when a project lacks a SPEC.md file or when the codebase is not organized in the required tasks/[task-id]/ directory structure, as the review process depends on these inputs.