Skip to content

Code Review Workflow

Reviews finished code changes for defects, risk, reuse gaps, and simplification before shipping.

Stars
40,512
Licence
MIT
Files
1 file

What it does

When installed, the agent acts as a reviewer for work that already exists. It reads a change against what it was meant to do, checks correctness first, then risk, reuse, simplification and test coverage, verifies each suspected problem before reporting it, and lists findings from most to least severe. For large changes it describes splitting the review into two separate passes (repo standards and intent) that are reported separately rather than merged into one ranking. It is advisory: it informs rather than blocks a release.

Great for

  • Developers who want a second look at a pull request or diff before merging
  • Teams that have coding standards documents (CLAUDE.md, AGENTS.md, CONTRIBUTING.md) they want applied consistently
  • Anyone who wants findings with file, line and a concrete failure scenario rather than generic advice

Skip it if

  • Writing or changing code itself, since the skill is explicitly review-only and never authors the change
  • Hard release or security gating, since the skill is advisory by default and does not fail builds (the SKILL.md says hard gates stay separate)
  • Non-code documents where there is no diff, intent source or standards file to compare against

Try asking

  • “Review my current branch against the ticket acceptance criteria and rank the findings by severity”
  • “Check this diff for security boundary and destructive operation risks before I merge”
  • “Look for code here that could be simplified or that duplicates existing utilities in the repo”
Step-by-step guide: how to use Code Review WorkflowInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    review.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

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

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip review.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 reviewer the task, spec or ticket text so the intent axis has something to check against
  • Keep the review in a fresh context rather than the session that wrote the change, as the skill requires separate lanes
  • Treat findings as candidates to verify; the skill itself says unreproducible findings are noise

What's inside

  • SKILL.md

Redistributed unmodified from Yeachan-Heo/oh-my-claudecode 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.

Find exactly what
you're looking for