agentsclimarketplace

Ui design

Skill kreek/consult/plugin/skills/ui-design

Use for frontend UI, layouts, components, responsive behavior, visual design, and usability.From its SKILL.md

Install
npx -y skills add kreek/consult --skill ui-design

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.9 KB, 847 tokens by cl100k_base, as published. Nobody here has run it

UI Design

Iron Law

START FROM THE USER TASK AND HIERARCHY; EVERY ELEMENT EARNS ITS PLACE.

When to Use

  • Building or changing any user-facing UI surface: layout, components, design systems, typography, color, motion, responsive behavior, or frontend structure. Simple forms, single-page apps, and "just basic styling" all qualify; the polished-product threshold is too high a bar.

When NOT to Use

  • Backend API shape; use api.
  • Accessibility-specific implementation or review; use accessibility.
  • Frontend runtime debugging or tests only; pair with proof and browser tooling.
  • Performance profiling beyond UI design choices; use performance.

Core Ideas

  1. One screen has one primary action and a clear information order.
  2. Use a small token system for spacing, type, color, radius, and motion. Apply tokens consistently; avoid stray one-off values.
  3. Accessibility is a design input, not a later review pass. Design keyboard, focus, contrast, reduced motion, touch targets, and screen-reader flow up front.
  4. Component APIs express intent and state, not implementation convenience.
  5. Modern CSS should reduce JavaScript and layout hacks when browser support allows it.

Workflow

  1. Identify the user, task, device constraints, and primary action. Choose existing framework/design-system patterns before inventing new ones.
  2. Define hierarchy, layout, states, empty/error/loading behavior, and responsive rules.
  3. Apply tokens consistently. Remove elements that do not improve comprehension, trust, or action.
  4. Verify with real rendering, keyboard navigation, contrast, and reduced-motion behavior.

Verification

  • Hierarchy survives a squint/blur test, and one primary action is visually dominant per screen.
  • Token system covers spacing, type, color, radius, motion; no stray one-off values in committed CSS.
  • Accessibility inputs were considered at design time, not patched after: keyboard reach, focus order, contrast, reduced motion, touch target, screen-reader flow.
  • UI states exist for loading, empty, error, disabled, and success where applicable.

Tripwires

TriggerDo this insteadFalse alarm
"Add a card/section so it looks richer"Start from the user task and hierarchy. Remove elements that do not improve comprehension, trust, or action.The card groups repeated peer items or frames a real tool surface.
"We'll add loading/error/empty states later"Define the required states with the layout.The component cannot load, fail, or be empty.
"One-off spacing fixes this screen"Use or extend the token system intentionally.A browser or platform quirk needs a documented local fix.
"Custom control first, semantics later"Start with native controls and route accessibility-specific behavior to accessibility.The existing design system control already proves semantics and keyboard behavior.
"Pick the frontend framework before the interaction model"Define the UI task, state, and flow first; use existing repo patterns where possible.The user explicitly asked to compare frameworks.

Handoffs

  • Use proof for UI behavior tests and browser-verified flows.
  • Use accessibility for WCAG, ARIA, semantic HTML, keyboard, focus, contrast, screen-reader, and inclusive-design work.
  • Use performance for measured Core Web Vitals or rendering regressions.
  • Use documentation for design-system usage docs and ADRs.

References

  • references/canon.md: design principles, product/tool defaults, common failure modes.
  • references/frameworks.md: frontend framework tradeoffs.
  • references/platforms.md: platform and government design systems.
  • references/css.md: modern CSS capabilities.
  • references/typography.md: Bringhurst-informed typography for product UI.

What ships with it: 6 files

33.3 KB alongside SKILL.md

agents/

references/

Gives 0 of the 12 instructions most design frontend skills give in 847 tokens

Counted across 1,169 of the 1,878 authors here whose files we hold, read 2026-08-07

  • Use CSS variables for color consistencyin 72 of 1169, across 23 files
  • Commit to one bold aesthetic direction before codingin 72 of 1169, across 27 files
  • Match implementation complexity to the aesthetic visionin 70 of 1169, across 20 files
  • Add atmospheric background effects and texturesin 57 of 1169, across 9 files
  • Use unexpected spatial compositions and layoutsin 56 of 1169, across 8 files
  • Implement real working codein 55 of 1169, across 7 files
  • Vary themes and aesthetics across different designsin 48 of 1169, across 7 files
  • Launch chromium in headless modein 47 of 1169, across 4 files
  • Close the browser when donein 47 of 1169, across 4 files
  • Run provided scripts with help flag firstin 47 of 1169, across 4 files
  • Wait for network idle statein 47 of 1169, across 4 files
  • Use descriptive selectors for elementsin 47 of 1169, across 4 files

Said here and by no other author read

  • start from user task and hierarchy
  • use a small token system for styling
  • choose existing framework patterns first
  • remove elements that do not aid comprehension
  • verify with real rendering and keyboard navigation
  • prefer modern CSS over layout hacks

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,679. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.