What problem does it solve? AI-generated website design references often collapse into repetitive defaults: left-text/right-image heroes, purple-blue gradients, cloned card sections, and single tall images that cram an entire page into one frame. This Skill enforces disciplined art direction so generated design comps look premium, varied, and directly implementable by developers or coding models. ## Core Features & Use Cases - One image per section: A landing page with 8 sections produces 8 separate horizontal images, never one collapsed tall slice, with sequential labeled output when only one image can render at a time. - Combinatorial variation engine: Picks theme paradigm, background mode, typography character, hero architecture, composition anchors, CTA styles, and a narrative concept spine, then commits to one consistent palette across all frames. - Anti-AI-slop rules: Bans generic copy ("unleash", "revolutionize"), fake brand names, gradient-text shortcuts, blob clutter, and default hero layouts, while enforcing spacing, rhythm, and conversion-aware section structure. - Use Case: Ask for a landing page for a premium SaaS product and receive 6-8 labeled section images (hero, trust bar, features, showcase, testimonials, pricing, CTA) that share one brand world and can be handed directly to a developer to recreate in code. ## Quick Start Generate a landing page design for my SaaS product with one horizontal image per section, using a cinematic dark theme and varied hero composition.
