What problem does it solve? LLM-generated frontend code tends to repeat the same layouts, produce narrow multi-line headings, leave empty gaps in bento grids, and ship static pages without motion. This Skill enforces a strict design engineering checklist so generated React/Tailwind pages follow editorial typography rules, AIDA page structure, and GSAP scroll animations. ## Core Features & Use Cases - Layout Randomization: Simulates a deterministic Python RNG step to select hero architectures, typography stacks, components, and GSAP paradigms so outputs vary between generations. - Grid and Typography Constraints: Enforces gapless bento grids via grid-flow-dense, wide H1 containers limited to 2-3 lines, and bans meta-labels like "SECTION 01". - GSAP Motion Rules: Requires ScrollTrigger patterns such as pinned sections, scrubbed text reveals, image scale/fade on scroll, and card stacking. - Use Case: Ask for a landing page for a SaaS product and receive a complete React/Tailwind page with a cinematic hero, dense bento feature grid, pinned scroll gallery, and a pre-flight design plan verifying spacing and contrast rules. ## Quick Start Use the gpt-taste skill to build a landing page for my product with a hero, bento features, and scroll animations.