What problem does it solve? AI-generated frontend code tends to fall back on the same templated aesthetics: purple gradients, centered heroes, three equal feature cards, and generic glassmorphism. This Skill forces the agent to read the brief first, infer the right design direction, and ship interfaces that do not look machine-generated. ## Core Features & Use Cases - Brief Inference and Design Read: Classifies page kind, vibe words, audience, and constraints before writing code, then declares a one-line design direction. - Three-Dial Configuration: Tunes DESIGN_VARIANCE, MOTION_INTENSITY, and VISUAL_DENSITY with signal-based inference tables and use-case presets. - Design System Mapping: Selects official packages (Fluent, Material, Carbon, Polaris, shadcn/ui, Tailwind v4) when the brief matches a real system instead of hand-rolling CSS. - Hard Layout and Accessibility Rules: Enforces hero viewport fit, eyebrow restraint, CTA contrast checks, palette locks, and a mechanical pre-flight audit. - Use Case: Ask the agent to redesign a premium consumer landing page; it will avoid the banned beige-and-brass default palette, pick a differentiated color family, generate real image assets, and pass a pre-flight checklist before shipping. ## Quick Start Redesign my SaaS landing page using the design-taste-frontend skill so it does not look like a generic AI template.