code-review

Review code changes and diffs with severity-ranked findings and minimal fix suggestions.

Updated Mar 25, 2026
One-click install
npx skills add https://github.com/Femad-6/my-skills --skill code-review-femad-6
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/Femad-6/my-skills/tree/main/.github/skills/code-review
Command: npx skills add https://github.com/Femad-6/my-skills --skill code-review-femad-6

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manual code review often misses critical bugs, security risks, and regression issues, or produces vague feedback that is hard to act on. This Skill structures the review process so every finding is severity-ranked, located, and paired with a minimal actionable fix. ## Core Features & Use Cases - Severity-First Findings: Classifies issues as Critical, High, Medium, or Low, each with file location, root cause, impact scope, and a minimal fix suggestion. - Regression Risk Analysis: Systematically checks boundary conditions, concurrency and async behavior, null and exception paths, and configuration compatibility. - Test Coverage Check: Verifies whether tests were added or updated and identifies missing critical test cases. - Use Case: Paste a git diff or list of changed files before merging a PR, and receive a structured report with findings, open questions, a change summary, and suggested tests. ## Quick Start Review this git diff for bugs, security risks, and missing tests, and list findings by severity.

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 pull request for bugs and security risks?

Provide the changed file list or git diff along with optional requirement notes. The review outputs findings ranked by severity, each including the file and line location, root cause, impact scope, and a minimal fix suggestion.

How to check if a code change has regression risk?

Regression risk is assessed by examining boundary conditions, concurrency and async behavior, null and exception paths, and configuration compatibility. The review explicitly flags these areas and lists remaining risks even when no blocking issues are found.

What output format does a structured code review produce?

The output contains four sections: findings sorted by severity, open questions and assumptions, a brief change summary, and suggested executable tests. This keeps conclusions verifiable and avoids subjective wording.

Can code review work without requirement documentation?

Yes, requirement notes and run instructions are optional inputs. When information is insufficient, the review lists its assumptions first before drawing conclusions, so results remain traceable.

What are the limitations of automated code review?

The review focuses on functional errors, security, performance, maintainability, and test gaps, but deliberately skips style-only details unless they harm readability or maintenance. It cannot execute code, so runtime behavior must be verified through the suggested tests.