Skip to content

Code Review (Standards and Spec)

Reviews your branch's changes against repo coding standards and the original spec, side by side.

Stars
309,658
Licence
MIT
Files
2 files

What it does

When you ask for a review of changes since a commit, branch, or tag, the agent checks the diff in two separate passes: whether the code follows the project's documented coding standards, and whether it does what the linked issue or spec asked for. It also flags common code smells (such as unclear names or duplicated logic) as judgement calls rather than errors. Results are shown as two separate reports so a pass on one axis cannot hide a problem on the other.

Great for

  • Developers reviewing a feature branch or pull request before merging
  • Teams that want a second look at work-in-progress changes against their written standards and issue requirements

Skip it if

  • Projects with no written spec or issue: the Spec check will be skipped, so it only covers standards and smells (less useful, not harmful)
  • Anyone expecting automated linting or tests to be run: the skill reviews by reading the diff and does not replace existing tooling

Try asking

  • “Review everything since main on this branch against our coding standards and the issue spec”
  • “Review the changes since commit a1b2c3d”
  • “Do a standards and spec review of my work-in-progress changes since the v2.0 tag”
Step-by-step guide: how to use Code Review (Standards and Spec)Install, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    code-review.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

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

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip code-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 an explicit fixed point (commit, branch, tag or main) so the diff is well defined.
  • Reference the issue number in commit messages so the agent can find the spec automatically.
  • Treat smell findings as prompts to consider, since they are labelled judgement calls.

What's inside

  • agents/openai.yaml
  • SKILL.md

Redistributed unmodified from mattpocock/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.