Ensure accessibility in UI components including semantic HTML, ARIA attributes, keyboard navigation, and WCAG 2.2 AA compliance.
中文简介正在准备中。当前显示原文说明。
Create distinctive, production-grade web interfaces with intentional aesthetic direction. Avoids generic AI aesthetics (centered layouts, Inter/Arial, purple gradients, uniform corners). Use for UI components, dashboards, landing pages, React/HTML/CSS layouts where design quality matters.
中文简介正在准备中。当前显示原文说明。
查看安装方法安装前可以查看交给 AI 代理的指令内容。
"Every design should feel specifically crafted for its context, never cookie-cutter."
This skill enables creation of distinctive, production-grade frontend interfaces that avoid the generic AI-generated aesthetic that has become ubiquitous. The goal is intentional design that reflects the specific context, brand, and purpose of the interface.
Generic AI-generated interfaces share predictable patterns that signal laziness:
Overused patterns to AVOID:
What to do instead:
/* High contrast, raw, confident */
:root {
--bg: #ffffff;
--text: #000000;
--accent: #ff0000;
--border: 2px solid #000000;
--radius: 0px;
}
.card {
border: var(--border);
padding: 2rem;
background: var(--bg);
}
Typography: Bold grotesques (Neue Haas Grotesk, Aktiv Grotesk, GT America) Motion: Sharp, instant transitions — no easing Layout: Rigid grid, full bleed, stark whitespace
/* Rich, layered, expressive */
:root {
--bg: #0a0a0f;
--text: #e8e3d9;
--accent: #c9a84c;
--accent-2: #7b5ea7;
}
.hero {
position: relative;
overflow: hidden;
background: radial-gradient(ellipse at 30% 50%, #1a0a2e, transparent);
}
Typography: Serif headlines (Playfair Display, Canela, GT Sectra) + tight monospace for data Motion: Cinematic reveals, staggered content entry, parallax depth Layout: Overlapping elements, diagonal breaks, generous negative space
/* Organic, crafted, human */
:root {
--bg: #faf8f4;
--text: #2c2416;
--accent: #c17b4a;
--accent-muted: #e8d5c0;
}
.surface {
background: var(--bg);
border: 1px solid rgba(0, 0, 0, 0.08);
box-shadow:
0 1px 3px rgba(0, 0, 0, 0.06),
0 4px 16px rgba(0, 0, 0, 0.04);
}
Typography: Humanist serifs (Lora, Freight Text) + friendly sans (Sora, Plus Jakarta Sans) Motion: Gentle, spring-based animations Layout: Asymmetric columns, hand-crafted feel
/* Dense, precise, functional */
:root {
--bg: #0d1117;
--text: #c9d1d9;
--accent: #58a6ff;
--grid: rgba(255, 255, 255, 0.04);
--mono: 'JetBrains Mono', 'Fira Code', monospace;
}
.data-panel {
background: rgba(13, 17, 23, 0.95);
border: 1px solid rgba(48, 54, 61, 0.8);
font-family: var(--mono);
}
Typography: Monospace throughout, tight line height, numerical data in tabular figures Motion: Real-time updates, pulsing activity indicators Layout: Dense grids, information hierarchy via size/weight/color
Don't reach for Inter, Arial, or system-ui. Choose fonts that carry meaning:
| Intention | Good Choices | Avoid |
|---|---|---|
| Premium / Editorial | Canela, GT Sectra, Freight Text | Georgia |
| Technical / Precise | JetBrains Mono, IBM Plex Mono | Courier |
| Modern / Professional | GT America, Neue Haas Grotesk | Inter, Helvetica |
| Warm / Approachable | Sora, Plus Jakarta Sans, Nunito | Roboto, Open Sans |
| Bold / Display | Clash Display, Cabinet Grotesk | Futura |
Use a clear hierarchy:
/* Display: hero headlines */
.text-display {
font-size: clamp(3rem, 8vw, 7rem);
font-weight: 900;
line-height: 0.95;
}
/* Heading: section titles */
.text-h1 {
font-size: clamp(2rem, 4vw, 3.5rem);
font-weight: 700;
line-height: 1.1;
}
/* Body: readable text */
.text-body {
font-size: 1rem;
line-height: 1.65;
letter-spacing: 0.01em;
}
/* Caption: secondary info */
.text-caption {
font-size: 0.8125rem;
line-height: 1.4;
letter-spacing: 0.04em;
text-transform: uppercase;
}
:root {
/* Dominant: 60% of UI */
--color-base: #f7f4ef;
--color-surface: #ffffff;
/* Secondary: 30% of UI */
--color-text: #1a1510;
--color-text-muted: #6b5e4e;
/* Accent: 10% of UI */
--color-accent: #c17b4a; /* primary actions */
--color-accent-sharp: #e05a2b; /* hover/active states */
}
The 60/30/10 rule prevents overwhelming accent usage.
High-impact reveals on page load (not scattered micro-interactions):
/* Staggered content entry */
@keyframes slideUp {
from {
opacity: 0;
transform: translateY(24px);
}
to {
opacity: 1;
transform: translateY(0);
}
}
.hero-heading {
animation: slideUp 0.6s cubic-bezier(0.22, 1, 0.36, 1) both;
}
.hero-sub {
animation: slideUp 0.6s cubic-bezier(0.22, 1, 0.36, 1) 0.1s both;
}
.hero-cta {
animation: slideUp 0.6s cubic-bezier(0.22, 1, 0.36, 1) 0.2s both;
}
/* State transitions: fast and responsive */
.button {
transition:
background-color 150ms ease,
transform 100ms ease;
}
.button:hover {
transform: translateY(-1px);
}
.button:active {
transform: translateY(0);
}
/* Panel reveals: deliberate */
.modal {
transition:
opacity 250ms ease,
transform 250ms cubic-bezier(0.34, 1.56, 0.64, 1);
}
For complex sequences:
Don't constrain everything to a uniform container:
/* Full-bleed accent with contained content */
.section-dark {
background: #0d1117;
/* No max-width here */
}
.section-dark .content {
max-width: 1200px;
margin: 0 auto;
padding: 0 2rem;
}
/* Let elements breathe */
.hero {
padding: 12rem 0 8rem;
}
.section {
padding: 8rem 0;
}
.card {
padding: 2.5rem;
}
/* 60/40 split instead of 50/50 */
.two-column {
display: grid;
grid-template-columns: 3fr 2fr;
gap: 4rem;
}
Before declaring a component done:
Before designing interfaces:
cat .claude/context/memory/learnings.md | grep -i "design\|ui\|frontend\|css"
After completing design work, record patterns:
.claude/context/memory/learnings.md.claude/context/memory/decisions.mdweb-artifacts-builder — Build React + Tailwind + shadcn/ui artifactsreact-expert — React component patternsstyling-expert — Tailwind CSS, CSS-in-JS patternsshadcn-ui — shadcn/ui component customizationenhance-prompt — Transform vague UI requests into design specs还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Ensure accessibility in UI components including semantic HTML, ARIA attributes, keyboard navigation, and WCAG 2.2 AA compliance.
中文简介正在准备中。当前显示原文说明。
Use when you want to improve response quality through meta-cognitive reasoning. Applies 15+ reasoning methods to reconsider and refine initial outputs.
中文简介正在准备中。当前显示原文说明。
N-round opposing-stance debates for trade-off analysis. Assigns pro/con roles to agents, runs structured debate rounds with quality scoring, and produces a moderator synthesis with confidence-rated recommendation. Generalizable to architecture, technology, security, and design decisions.
中文简介正在准备中。当前显示原文说明。
Force adversarial code review stance that eliminates confirmation bias — reviewer must find issues or re-analyze
中文简介正在准备中。当前显示原文说明。
Creates specialized AI agents on-demand when no existing agent matches a request. Use when the Router cannot find a suitable agent for a task. Enables self-evolution by generating persistent agents.
中文简介正在准备中。当前显示原文说明。
LLM-as-judge evaluation framework with 5-dimension rubric (accuracy, groundedness, coherence, completeness, helpfulness) for scoring AI-generated content quality with weighted composite scores and evidence citations
中文简介正在准备中。当前显示原文说明。