What problem does it solve? Frontend interfaces often ship with bland visuals, inconsistent design systems, accessibility gaps, and unverified quality. This Skill gives an AI agent a structured design-director workflow to build, evaluate, and refine UI so every surface meets a defined craft floor before shipping. ## Core Features & Use Cases - 23 routed commands: build (shape, init, document, extract), evaluate (critique, audit), refine (polish, bolder, quieter, distill, harden, onboard), enhance (animate, colorize, typeset, layout, delight, overdrive), and fix (clarify, adapt, optimize), plus a live browser variant mode. - Scored evaluation reports: critique runs dual independent assessments with Nielsen heuristic scoring and persona red flags; audit scores accessibility, performance, theming, responsive design, and implementation integrity with P0-P3 severity findings. - Web and native support: dedicated references for iOS and Android platform conformance, responsive adaptation, and a craft-floor checklist that bans common design anti-patterns. - Use Case: Ask the agent to critique a landing page; it runs a design review plus a deterministic detector scan, returns a scored report with prioritized fixes, then applies them via polish, layout, and typeset commands. ## Quick Start Ask the agent to run the impeccable critique command on your homepage file to get a scored design review with prioritized fixes.