Css Systems logo

Css Systems

Organization
staple-lab
css-systems

Use when wiring design tokens into a styling system or writing component styles - Tailwind v4, CSS Modules, StyleX, vanilla-extract, Panda CSS, Emotion - or when feeding tokens into a pre-styled component layer's theming (HeroUI, MUI, Mantine, Chakra, antd). Covers how tokens reach components in each system, variant patterns, theming, RSC compatibility, and how a design system ships its CSS to consumers.

Overview

Publisherstaple-lab
Repositorydesign-system-skills
Skill namecss-systems
Stars
7
Forks
0
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by staple-lab on GitHub. Read the source before you install it.

Installation

Install the Css Systems AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/staple-lab/design-system-skills/tree/main/plugins/design-system/skills/css-systems
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/staple-lab/design-system-skills.git /tmp/design-system-skills
mkdir -p .claude/skills
cp -r /tmp/design-system-skills/plugins/design-system/skills/css-systems .claude/skills/css-systems
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Css Systems in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Css Systems on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Css Systems is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

CSS systems

The CSS system's job is narrow: make tokens reachable from components, idiomatically. It does not define values — the DTCG token layer does, and tokens/build.mjs emits whatever artifact this system needs. Keeping that boundary is what lets you change CSS systems in three years without redesigning anything.

Read the chosen system from design-system.config.json → stack.cssSystem.

The universal substrate

Whatever the system, the generated tokens.css is always emitted and always loaded first. Every other artifact points back at those custom properties rather than inlining values, which is what makes [data-theme] switching work at runtime with no rebuild and no flash.

css
@import './tokens/dist/tokens.css';   /* always, in every system */

Tailwind v4

CSS-first. @theme inline maps Tailwind's namespaces onto the token vars, so utilities re-theme for free:

css
/* tokens/dist/theme.css — generated */
@import 'tailwindcss';
@import './tokens.css';

@theme inline {
  --color-background: var(--ds-color-bg-default);
  --color-accent: var(--ds-color-bg-accent);
  --color-accent-foreground: var(--ds-color-fg-on-accent);
  --spacing-4: var(--ds-space-4);
  --radius-md: var(--ds-radius-md);
  --ease-emphasized: var(--ds-easing-emphasized);
}

inline is the important keyword: without it Tailwind resolves the variable at build time and freezes the light-theme value into the utility, and dark mode silently stops working. With it, the utility keeps pointing at the live custom property.

Variants belong in CVA (class-variance-authority) or tailwind-variants, not in ad-hoc template strings:

tsx
const button = cva('inline-flex items-center justify-center rounded-md font-medium transition-colors ' +
                   'focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-ring focus-visible:ring-offset-2 ' +
                   'disabled:pointer-events-none disabled:opacity-50', {
  variants: {
    variant: {
      primary:   'bg-accent text-accent-foreground hover:bg-accent-hover active:bg-accent-active',
      secondary: 'bg-subtle text-foreground hover:bg-subtle-hover',
      ghost:     'bg-transparent text-foreground hover:bg-subtle',
      danger:    'bg-danger text-danger-foreground hover:bg-danger-hover',
    },
    size: { sm: 'h-8 px-3 text-sm', md: 'h-10 px-4 text-sm', lg: 'h-12 px-5 text-base' },
  },
  defaultVariants: { variant: 'primary', size: 'md' },
});

Merge consumer classes with tailwind-merge (cn()), or className="p-8" loses to the component's own p-4 by source order rather than by intent — a bug that looks like the escape hatch is broken.

Shipping to consumers: publish the compiled CSS and a @source directive pointing at your package, so consumers' Tailwind scans your components for classes. Document which; getting this wrong produces a component library that renders unstyled in someone else's app.


CSS Modules

The most durable option. Tokens are already custom properties, so there is nothing to wire:

css
/* Button.module.css */
.button {
  display: inline-flex; align-items: center; justify-content: center;
  height: var(--ds-control-height-md);
  padding-inline: var(--ds-control-padding-x-md);
  border-radius: var(--ds-control-radius);
  font: var(--ds-type-label-font-weight) var(--ds-type-label-font-size) / var(--ds-type-label-line-height) var(--ds-type-label-font-family);
  transition: background-color var(--ds-duration-fast) var(--ds-easing-standard);
}
.button:focus-visible {
  outline: var(--ds-focus-ring-width) solid var(--ds-color-ring);
  outline-offset: var(--ds-focus-ring-offset);
}
.primary { background: var(--ds-color-bg-accent); color: var(--ds-color-fg-on-accent); }
.primary:hover:not(:disabled) { background: var(--ds-color-bg-accent-hover); }

Compose variants with a small local helper rather than a dependency:

tsx
import s from './Button.module.css';
const cx = (...c: (string | false | undefined)[]) => c.filter(Boolean).join(' ');
<button className={cx(s.button, s[variant], s[size], className)} />

Shipping: one stylesheet, consumers import '@acme/ui/styles.css'. The simplest consumer story of any system, which is a real argument for it in a library.


StyleX

Build-time atomic CSS with typed vars and — the genuine differentiator — deterministic merge order. The last style applied wins, predictably, instead of depending on CSS source order and specificity:

ts
import * as stylex from '@stylexjs/stylex';
import { tokens } from '../tokens/dist/tokens.stylex';

const styles = stylex.create({
  base: {
    display: 'inline-flex',
    height: tokens.controlHeightMd,
    borderRadius: tokens.controlRadius,
    transitionDuration: tokens.durationFast,
  },
  primary: {
    backgroundColor: { default: tokens.colorBgAccent, ':hover': tokens.colorBgAccentHover },
    color: tokens.colorFgOnAccent,
  },
});

<button {...stylex.props(styles.base, variant === 'primary' && styles.primary)} />

The trade-off: consumers cannot restyle with className from outside, by design. That is a feature in an app and a problem in a distributed library — expose a style prop with token-typed values instead, and be explicit about it in the docs.


vanilla-extract

Typed styles in .css.ts, and the strongest correctness story: a theme that forgets a token is a type error.

ts
// Button.css.ts
import { style, styleVariants } from '@vanilla-extract/css';
import { vars } from '../tokens/dist/contract.css';

export const base = style({
  display: 'inline-flex',
  height: vars.control.height.md,
  borderRadius: vars.control.radius,
  ':focus-visible': { outline: `2px solid ${vars.color.ring}`, outlineOffset: '2px' },
});

export const variant = styleVariants({
  primary: { background: vars.color.bg.accent, color: vars.color.fg.onAccent },
  ghost:   { background: 'transparent', color: vars.color.fg.default },
});

createGlobalThemeContract (what the token build emits) points the contract at the existing custom properties, so you keep runtime theme switching and compile-time safety. recipes covers variants with typed props.


Panda CSS

Token-first config with excellent variant ergonomics via recipes:

ts
import preset from './tokens/dist/preset';
export default defineConfig({ presets: ['@pandacss/dev/presets', preset], /* ... */ });
ts
export const button = cva({
  base: { display: 'inline-flex', h: 'control.height.md', rounded: 'control.radius' },
  variants: { variant: { primary: { bg: 'color.bg.accent', color: 'color.fg.on-accent' } } },
});

Slot recipes are the cleanest multi-part-component styling of any system here. Cost: a real build step, and a smaller community when something goes wrong.


Emotion / styled-components

Runtime CSS-in-JS. Do not start here in 2026 — both need a client runtime, which fights React Server Components and streaming SSR, and styled-components is in maintenance. Support them only for migration. If you are stuck with one, drive it from the same custom properties (background: var(--ds-color-bg-accent)) so the migration path stays open.


Pre-styled component layers

When design-system.config.json → stack.componentLayer names a pre-styled library (heroui, mui, mantine, chakra, antd, radix-themes, park), this skill's job changes: you are not styling components — the vendor did that — you are feeding the vendor's theming surface from the tokens. The full per-library wiring (and which of it is verified vs. needs checking) lives in ${CLAUDE_PLUGIN_ROOT}/skills/design-system-architect/references/stack-constraints.md; the shape of the work:

  • HeroUI (Tailwind v4 required — npm-enforced peer dep): CSS-side. Assign HeroUI's semantic variables from the generated --ds-* custom properties in :root, [data-theme='light'] and .dark, [data-theme='dark'] blocks — --accent: var(--ds-color-bg-accent), --accent-foreground: var(--ds-color-fg-on-accent), --background, --foreground, and so on. HeroUI's own @theme inline bridge (themes/shared/theme.css) then carries those into Tailwind utilities, and its color-mix()-calculated hover/soft/field variants derive automatically — one assignment re-themes the interactive states too. Use the same data-theme values as the token build so one toggle flips both layers.
  • MUI / Mantine / Chakra / antd: JS-side. Token values flow into createTheme / MantineProvider / the Chakra system config / ConfigProvider theme.token — as literal values from a token-build artifact, not var(--ds-*) strings, because these engines compute derived colors (hover tints, contrast text) from the values and cannot compute from an unresolved var().

Either way the token layer stays the source of truth; what changes is the artifact the build emits and where it is handed over. The tokens.css substrate is still loaded first — product code outside the vendor's components consumes --ds-* directly, as everywhere else in this skill.

  1. Semantic tokens only. No raw hex, no magic px. New value → new token first. This is the lint layer's main job.
  2. State from data-*, not from React. [data-state='open'], [data-disabled]. The DOM is already the source of truth.
  3. :focus-visible, never :focus. :focus fires on mouse click, which is what gets focus styles deleted, which is what strands keyboard users.
  4. Logical properties. padding-inline, margin-block, inset-inline-start. Free RTL, and it costs nothing to type.
  5. One escape hatch, deliberately chosen. className passthrough, a style prop, or exposed component-level CSS vars. Without one, people fork the component and you lose them permanently.
  6. No !important inside the system. If you need it, the specificity design is wrong.
  7. Container queries over media queries for component-level responsiveness — a component should respond to its space, not the viewport's. @container (min-width: 24rem) is what makes a card work in both a sidebar and a full-width grid without a prop.

Frequently asked questions

What does the Css Systems AI skill do?

Use when wiring design tokens into a styling system or writing component styles - Tailwind v4, CSS Modules, StyleX, vanilla-extract, Panda CSS, Emotion - or when feeding tokens into a pre-styled component layer's theming (HeroUI, MUI, Mantine, Chakra, antd). Covers how tokens reach components in each system, variant patterns, theming, RSC compatibility, and how a design system ships its CSS to consumers.

Why use Css Systems on TypingMind?

Because you install it once and use it with any model. Css Systems is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Css Systems in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/staple-lab/design-system-skills/tree/main/plugins/design-system/skills/css-systems. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Css Systems?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Css Systems?

As many as you like. As long as a model supports skills, you can use Css Systems with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Css Systems AI skill free?

Yes. It is published on GitHub by staple-lab under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇