frontend design

Design Community

Skill notes

Skill Activation & Context

Load this skill only when the task explicitly touches UI/frontends to keep the base prompt lean. Reference the skill name when suggesting it to other agents. Before designing, collect the minimum viable brief: product type, audience, success metric, preferred frameworks, and any brand/visual cues (fonts, palette, motion, background texture). If the user supplies no strong direction, offer 2–3 curated aesthetics (e.g., monochrome editor, warm editorial, glassmorphism dashboard, retro terminal) and ask them to pick or blend.

Typography Systems

Always define a hierarchy: display, headline, body, mono/utility. Provide the exact font family plus a safe fallback stack. Map font weights, letter spacing, and line-height to token names in CSS variables so the runtime can apply them anywhere. Include one expressive font pairing suggestion (e.g., Inter + IBM Plex Mono) whenever the user only specifies “modern” or “futuristic”.

Color Direction

Deliver a compact palette: primary, secondary, accent, background, surface, text, success, warning, danger. Call out gradients, glass layers, or duotone overlays if they reinforce the theme. For dark themes, note the exact luminance steps to avoid muddy grays. Provide contrast guidance (WCAG AA/AAA) and call out which colors to reserve for CTA vs. supporting elements.

Motion & Interaction Energy

Specify durations (e.g., 120ms micro, 240ms macro), easing curves, and a small set of reusable transitions (slide, fade, scale, blur). Describe micro-interactions for hover/focus/active including motion + color shifts to reinforce tactility. Note whether scroll-triggered reveals, parallax, or staggered animations are desirable for the requested experience.

Background & Depth Treatments

Describe the canvas: noise overlays, gradient types, subtle illustrations, or photography directions. If depth is desired, prescribe elevation tiers (cards, modals, floating toolbars) and the matching shadow/blur token for each. Call out when to use split planes, layered blobs, or tinted glass to break away from the “single white card” default.

Fluxly profile Runtime and permissions are inferred by Fluxly static checks, not author declarations or runtime test results.
Domain
Design
Author-declared agents
No explicit declaration found; this is not inferred or tested compatibility
Static check
88 / 100 · heuristic scan, not runtime safety proof
Runtime evidence
Not run; no isolated runtime test evidence is currently recorded
Author / version / license
@0-CYBERDYNE-SYSTEMS-0 · no license declared
Token usage
Lean
Setup complexity
Guided setup
External API key
Not required
Operating systems
Unspecified (assume cross-platform)
Runtime requirements
No explicit requirements detected
Detected file/system behavior
  • Read-only
  • Write / modify
  • Shell exec
Network behavior
Local-only
Install commands
None (reference only)

Heads up: Tool permissions are not declared; actual access depends on the runtime.

Output preview frontend-design-0-cyberdyne-systems-0.preview
# Rich Output Guidance

- Prefer multi-file or component-driven output (React/Vite/Tailwind, Next.js, Astro, etc.) whenever available; call the `component_library` or other code-focused skills to scaffold file trees before writing code.
- When artifacts must be a single HTML file, still modularize via `<template>` blocks or clearly documented sections for future extraction.
- Describe the layout skeleton (sections, component tree, responsive breakpoints) before writing code so advanced builders (web artifacts tool, code execution) can hydrate it.
- Include QA cues from the skill (accessibility, performance budgets) in every handoff to maintain the deterministic improvements described in the Claude blog article.

Discussion

Powered by GitHub Discussions. Sign in with GitHub to comment, react, or subscribe.