What problem does it solve? LLM-generated frontend code tends to fall back on the same templated aesthetics: AI-purple gradients, centered heroes, three equal feature cards, and default Inter typography. This Skill forces the agent to read the brief first, infer the right design direction, and ship interfaces that look deliberately designed rather than auto-generated. ## Core Features & Use Cases - Brief Inference and Design Read: Analyzes page kind, vibe words, references, audience, and brand assets to produce a one-line design read before writing any code, with CI-mode defaults when no human can clarify. - Three-Dial Configuration: Tunes DESIGN_VARIANCE, MOTION_INTENSITY, and VISUAL_DENSITY via inference tables and use-case presets to drive layout, motion, and density decisions. - Design System Mapping and Anti-Slop Rules: Selects official design systems (Fluent, Material, Carbon, shadcn/ui, Tailwind v4) when appropriate, and enforces hard pre-flight rules on typography, color, hero layout, eyebrows, CTAs, contrast, and image assets. - Use Case: Given a brief for a premium consumer cookware landing page, the Skill derives a non-default palette (avoiding the banned beige-and-brass family), picks an asymmetric hero layout, generates real image assets, and passes a mechanical pre-flight audit before shipping. ## Quick Start Use the taste-skill to design and build a landing page for my SaaS product based on this brief, inferring the design direction and running the full pre-flight check before shipping.
