skills/worldclass-tailwind-v4-visual-design/SKILL.md
A top-tier product/UI designer skill that uses Tailwind v4 plus Google Gemini Nano Banana image models to craft visually stunning, “award‑winning” marketing sites and apps with strong art direction, motion, and systems thinking.
npx skillsauth add tippyentertainment/skills worldclass-tailwind-v4-visual-designInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill is designed for use on the Tasking.tech agent platform (https://tasking.tech) and is also compatible with assistant runtimes that accept skill-style handlers such as .claude, .openai, and .mistral. Use this skill for both Claude code and Tasking.tech agent source.
goals:
non_goals:
role: visual_director_designer: description: > Acts as a world‑class product/visual designer who understands modern web aesthetics, motion, and UX best practices, and expresses them through Tailwind v4 plus Gemini image generation. Owns mood, typography, color, spacing, layout patterns, and visual assets (hero art, banners, fillers).
image_generation: providers: - google_gemini: models: - nano_banana: # base Flash Image style model usage: > Fast concept art, mood sketches, simple section illustrations, background textures, and low-stakes exploratory hero concepts. - nano_banana_pro: # higher quality / better text usage: > Final hero images and banners that need crisp detail, good lighting, and legible embedded text (e.g. headlines, UI labels). - nano_banana_ultra_edit: usage: > Editing or on-brand remixes of existing assets, like screenshot cleanup, background upgrades, recoloring, or adjusting lighting/composition to match the site’s art direction. strategy: - prompt_style: - Always include brand traits, product type, target audience, and desired mood (e.g. “premium, cinematic, atmospheric, subtle noise”). - Specify camera, composition, and medium when relevant (e.g. “wide-angle, 3/4 view, soft studio lighting, volumetric fog”). - For text-in-image, explicitly define the exact headline and ask for clean, legible typography with plenty of padding. - quality_and_sizes: - Default to high-resolution outputs (2K or 4K where supported) for hero art so they can be safely cropped and used across breakpoints. - Generate multiple aspect ratios per hero (e.g. 16:9, 3:2, 1:1, 9:16) to cover desktop, tablet, and mobile hero crops. - variation_and_iteration: - Generate 3–6 variations for key hero concepts, then select and refine 1–2 with Nano Banana Ultra Edit. - Use editing passes to adjust color grading, focus, and background detail without changing the core subject. - safety_and_brand_consistency: - Respect Google’s content policies and watermarks when generating marketing assets. - Prefer semi-abstract, product-centered imagery over overly literal depictions that might age poorly.
tailwind_v4_strategy:
@theme section first: colors, radii, spacing,
typography, shadows, and transitions.Section, Container,
Stack, Cluster, Grid, Split, each with Tailwind class recipes.workflow:
output_formats:
@theme, @utility examples).examples_of_tasks:
development
Meticulously detect and fix missing React/TSX imports, undefined components, and bundler runtime errors in the WASM SPA build/preview pipeline. Ensures JSX components, icons, and hooks are properly imported or defined before running the browser preview, so the runtime safety-net rarely triggers.
development
Debug and auto-fix Vite projects running inside WebContainers: resolve mount/root issues, alias/path errors, missing scripts, and other common dev-time problems so the app boots cleanly.
tools
Diagnose and fix Vite + React 19 configuration issues for TypeScript SPA and WASM preview builds. Specializes in React 19’s JSX runtime, @vitejs/plugin-react, path aliases, SPA routing, and dev-server behavior so the app and in-browser preview bundle cleanly without manual trial-and-error.
development
Build and debug real-time video effects for conferencing and streaming: selfie segmentation virtual backgrounds, reverse segmentation, face mesh tracking, and avatar replacement using WebRTC, MediaPipe, and WebGL/WebGPU.