review

Review code changes for correctness, security, style, architecture, and testing issues.

Updated Feb 20, 2026

Search Tags:#git #testing #diff analysis #architecture #code review #security

Ready to run

npx skills add https://github.com/supertyrelle/pelley --skill review-supertyrelle

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review
Source: https://github.com/supertyrelle/pelley/tree/main/skills/review
Command: npx skills add https://github.com/supertyrelle/pelley --skill review-supertyrelle

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you perform a disciplined code review so you can catch bugs, security risks, design drift, and missing tests before changes are merged.

Core Features & Use Cases

Multi-dimension review: Evaluates changes across correctness, security, style, architecture, and testing.
Scope-aware analysis: Handles staged changes, commits, commit ranges, pull requests, and file-specific reviews.
Actionable findings: Produces severity-ranked issues with concrete evidence and merge-ready guidance.

Quick Start

Review the staged changes in this repository and return prioritized findings with evidence, severity, and a clear merge verdict.

Frequently Asked Questions about review

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

FAQPage Schema

How do I check staged git diffs for security and correctness issues before merging?▼

To check staged git diffs, the review process evaluates code changes across correctness, security, style, architecture, and testing dimensions. It analyzes full file context and traces code paths to produce severity-ranked findings with actionable merge guidance.

What is the best way to perform an automated code review on a pull request?▼

The best way to perform an automated code review on a pull request is using scope-aware diff analysis. This evaluates staged changes or commit ranges, classifies findings by severity with concrete evidence, and returns a clear merge verdict.

Can I review file-specific git changes for architecture and testing gaps?▼

Yes, you can review file-specific git changes. The analysis applies to individual file modifications, reading the full file context to identify design drift, architecture issues, and missing tests before producing a final merge verdict.

Does code review analysis work with commit ranges or only staged changes?▼

Code review analysis works with both staged changes and commit ranges. It evaluates the full scope of provided changes, traces code paths to classify severity, and generates actionable evidence for pre-merge quality checks.

How do automated pre-merge quality checks identify missing tests in a diff?▼

Automated pre-merge quality checks identify missing tests by reading full file context and tracing code paths within the diff. The review classifies these testing gaps by severity and provides actionable evidence in the final merge verdict.

What limitations should I expect when reviewing code changes for security and style?▼

When reviewing code changes, limitations depend on providing full file context. The analysis requires reading complete files to trace code paths and classify security or style findings accurately; incomplete context may reduce the accuracy of the merge verdict.