What problem does it solve? AI-generated interfaces tend to converge on generic, recognizable patterns (gradient text, cream backgrounds, card grids, eyebrow labels) and ship with contrast, accessibility, and responsive defects. This Skill provides a disciplined design workflow that produces distinctive, on-brand, production-ready frontend code instead of AI slop. ## Core Features & Use Cases - 23 routed sub-commands: Build (craft, shape, init, document, extract), Evaluate (critique, audit), Refine (polish, bolder, quieter, distill, harden, onboard), Enhance (animate, colorize, typeset, layout, delight, overdrive), Fix (clarify, adapt, optimize), and Iterate (live browser-based variant picking). - Register-aware design rules: Separate brand and product references with font reflex-reject lists, color strategies (Restrained through Drenched), motion budgets, and absolute bans on saturated AI patterns. - Project context system: Scripts generate PRODUCT.md/DESIGN.md context, seed brand palettes in OKLCH, detect anti-patterns in the working tree, and recommend next commands from project signals. - Use Case: Run /impeccable audit on a dashboard to get a scored report across accessibility, performance, theming, responsive design, and anti-patterns, then follow the prioritized /impeccable polish and fix commands it recommends. ## Quick Start Ask the AI to run /impeccable init to set up project context, then invoke a command like /impeccable critique landing page to review a surface.