code-review

Review PRs, branches, or files with file:line references.

36|4|Updated Feb 11, 2026
One-click install
npx skills add https://github.com/xg-gh-25/SwarmAI --skill code-review-xg-gh-25
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/xg-gh-25/SwarmAI/tree/main/backend/skills/s_code-review
Command: npx skills add https://github.com/xg-gh-25/SwarmAI --skill code-review-xg-gh-25

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Structured, opinionated code review that surfaces actionable findings with file:line references and no external dependencies.

Core Features & Use Cases

  • Actionable findings with file:line references across modified files and diffs.
  • Deterministic review workflow covering correctness, security, design, readability, testing, and performance.
  • PR and file scope: review PR diffs, branches, or individual files; generate guidance and suggested improvements.
  • No external dependencies: operates using existing dev tooling and the agent's own reasoning to produce a consistent report.

Quick Start

Ask me to review a PR, branch, or file by saying 'review PR #123' or 'review this directory', and I will generate a structured report.

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 code quality and security issues?

Review a pull request by analyzing git diffs and touched files to surface actionable findings with specific file:line references. This structured code review process evaluates correctness, security, design, readability, testing, and performance across modified files.

Can I do a structured code review on a local Git branch without external dependencies?

Yes, you can perform a structured code review on a local Git branch without external dependencies. The review operates using built-in Git and Gh tooling alongside agent reasoning to analyze commits and produce a standardized, actionable report.

What is the best way to get actionable findings from a git diff analysis?

The best way to get actionable findings from a git diff analysis is applying a deterministic review workflow. This generates structured reports with precise file:line references and suggested improvements based on established review guidelines across multiple code dimensions.

Does static analysis work for reviewing individual files in a GitHub workflow?

Static analysis works for reviewing individual files in a GitHub workflow by analyzing touched files and diffs. It generates guidance and suggested improvements with file:line references, operating self-contained using existing dev tooling and agent-based analysis.

What are the limitations of agent-based code review compared to dedicated static analysis tools?

Agent-based code review limitations include relying on the agent's own reasoning rather than executing dedicated static analysis binaries. It operates without external dependencies using built-in Git tooling, which may limit language-specific deep semantic analysis compared to specialized linters.