Skip to content
Get packs

Code Reviewer

Reviews pull requests, diffs and snippets across many languages for security, async, error and performance issues, then gives a merge verdict.

Source Download skill
Repository
alirezarezvani/claude-skills
Stars
29,149
Licence
MIT
Files
15 files

What it does

When you ask an AI agent to review code, this skill gives it a structured checklist: general rules plus a rule file for the programming language in question. The agent looks for security problems such as SQL injection and missing authentication, resource and error-handling mistakes, and performance issues. It can also run bundled Python scripts that estimate a pull request's risk and complexity, score code quality, and produce a verdict (approve, request changes, or block) with blocking and non-blocking findings.

Great for

  • Developers who want a consistent first-pass review of a pull request or diff before a human reviews it
  • Teams working across several languages who want the same review standards applied to each
  • Anyone who wants a quick check of a single function or file for obvious security or error-handling problems

Skip it if

  • Replacing a human security audit or compliance review, because the checks are heuristic pattern rules and can miss issues or flag false positives
  • Auto-merging or gating production deployments on the verdict alone, since the scoring thresholds are fixed and not validated against your codebase

Try asking

  • “Review this PR and tell me if it is safe to merge: [paste diff]”
  • “Look over this Go handler and point out security or error-handling problems”
  • “Run a code review on the commit abc123 in my repo and list blocking issues”
  • “Check this Python function for SQL injection or missing auth”
Step-by-step guide: how to use Code ReviewerInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 15 original files, the MIT licence and an attribution note.

    code-reviewer.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip code-reviewer.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip code-reviewer.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Give the agent the full diff or the file path rather than a summary, so the language rule file and scripts can be applied correctly
  • Treat the verdict as a starting point and check any blocking finding yourself before acting on it
  • Run the scripts with --json if you want to integrate results into your own CI, and review the script source first since it executes locally

What's inside

  • expected_outputs/sample_c_clean_quality.json
  • expected_outputs/sample_c_smells_quality.json
  • expected_outputs/sample_csharp_clean_quality.json
  • expected_outputs/sample_csharp_smells_quality.json
  • expected_outputs/sample_java_clean_quality.json
  • expected_outputs/sample_java_smells_quality.json
  • languages/c.md
  • languages/cpp.md
  • languages/csharp.md
  • languages/dart.md
  • languages/go.md
  • languages/java.md
  • languages/kotlin.md
  • README.md
  • SKILL.md

Redistributed unmodified from alirezarezvani/claude-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.