Skip to content
Get All-Access
All skills
ProAI agents

Reviewer. Reviews a pull request or page for design-system drift and accessibility — findings ranked, each with a patch.

A Claude Code agent that runs the token check, the motion audit, the accessibility checks and the screen lint on what changed, merges the findings by cause and returns them ranked with a patch each.

Get All-AccessSee the workflow
Workflow

What it actually does. Step by step, in the order it runs.

  1. 1

    Scope

    The changed files against the base branch, or the paths and URL you name.

  2. 2

    Measure

    Token check, motion audit, accessibility (static, and the browser run when a URL is available), screen lint.

  3. 3

    Merge and rank

    By cause: accessibility first, then token drift that breaks dark mode, then motion.

  4. 4

    Report

    Counts per check, the top findings with file:line and a patch, and what could not be checked.

  5. Skills it runs

Example run

What a run looks like. An illustration of the summary it hands back.

You ask

“Review this branch.”

Or in other words

  • “Review this pull request for WunderUI drift.”
  • “Check the new onboarding screens before release.”
tokens 3 · motion 2 · a11y 1 (IconButton without aria-label) · screen lint 0 — 6 patches proposed
Install

Comes with WunderUI Pro. One folder for your agent.

The WunderUI Pro skills zip comes with your order (confirmation e-mail and Polar customer portal). Unzip it and copy the folder:

# from the WunderUI Pro skills zip
cp wunderui-skills/agents/wunderui-reviewer.md ~/.claude/agents/
cp -r wunderui-skills/skills/* ~/.claude/skills/

Claude Code loads it by itself when your request fits. Cursor, Codex and other agents: point them at the skill's .

What it needs

  • Claude Code
  • Node.js 22+ and Chrome for the browser checks