code-review

Review code diffs for correctness, security, performance, and maintainability.

110|14|Updated May 12, 2026
One-click install
npx skills add https://github.com/EliasOulkadi/shokunin --skill code-review-eliasoulkadi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/EliasOulkadi/shokunin/tree/main/.pack/skills/code-review
Command: npx skills add https://github.com/EliasOulkadi/shokunin --skill code-review-eliasoulkadi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you review code changes quickly and reliably, reducing the risk of correctness bugs, security vulnerabilities, and maintainability problems before they land.

Core Features & Use Cases

  • Diff-first review workflow: Accepts pasted diffs, commit hashes, or git ranges to establish an accurate scope of changes.
  • Context-aware analysis: Reads full changed files and checks dependencies/callers to avoid shallow, misleading feedback.
  • Actionable, prioritized findings: Produces a structured review with severity levels (P0/P1 fixes required, P2 suggestions), plus guidance tailored to the user’s stated requirements.

Use case: reviewing a pull request that modifies an authentication boundary, validate error handling and authorization checks using the diff as the single source of truth, then request changes only for P0/P1 findings.

Quick Start

Ask the AI to review your changes by pasting the git diff or providing a commit range, and include any requirements that motivated the change.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I review a git diff for security vulnerabilities and bugs?

To review a git diff for security vulnerabilities and bugs, paste the diff or provide a commit range to establish the scope, then analyze correctness, security, performance, and maintainability issues using the changed files as context.

What is the best way to automate pull request reviews for maintainability?

Automating pull request reviews for maintainability involves analyzing the input diff, gathering context across affected files and dependencies, and producing prioritized findings structured by severity levels like P0 or P1.

Can I use a commit hash to check code changes for performance issues?

Yes, you can use a commit hash to check code changes for performance issues by extracting the diff from the specified git commit and evaluating the modified implementation behavior against your stated requirements.

Does code review work with git ranges and requirement-driven validation?

Code review works with git ranges by extracting changes across the specified range, gathering context from affected files, and validating modified implementation behavior against your specific requirements to ensure correctness.

How do I prioritize feedback when reviewing a code diff?

Prioritize feedback when reviewing a code diff by classifying findings into severity levels from P0 to P3, requesting changes only for critical P0 and P1 fixes while treating P2 items as suggestions.

Why does shallow diff analysis miss correctness bugs in modified files?

Shallow diff analysis misses correctness bugs because it lacks the context of full changed files and their callers, leading to misleading feedback instead of accurate severity classification and requirement-driven validation.