Shain Dls logo

Shain Dls

Community
Dishain
shain-dls

Token-based design system skill for Figma + code handoff. Use whenever working with Figma components, design tokens, color variables, token migration, component creation, design-to-code handoff, or DS audits. Trigger on: design system, DS, DLS, tokens, figma variables, component spec, token naming, token hierarchy, which color to use, variant properties, theme config, dark mode. Even for "which color should I use" or "create a component" — use this skill.

Overview

PublisherDishain
RepositoryShain-Design-System-Skill
Skill nameshain-dls
Stars
8
Forks
0
Bundled files
2
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.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Dishain on GitHub. Read the source before you install it.

Installation

Install the Shain Dls 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/Dishain/Shain-Design-System-Skill/tree/main
  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/Dishain/Shain-Design-System-Skill.git \
  .claude/skills/shain-dls
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Shain Dls 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 Shain Dls 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 Shain Dls 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.

Shain Design Language System (DLS)

A token-based design system for product UI work. Components use semantic tokens only — no hardcoded colors, spacing, or typography. Figma is the source of truth; code is generated or guided by Figma exports.

For the full Figma best practices reference, read references/figma-best-practices.md. For guidance on evolving the token system, read references/token-improvements.md.


1. Architecture

Three-tier token hierarchy

Primitive → Semantic → Component-specific (optional)

colors/blue/500          → #0097DB
action/primary/default   → {colors/blue/500}
button/bg/primary        → {action/primary/default}
  • Primitive — raw palette values (colors/blue/500, spacing/4, radius/md). Never referenced directly from components.
  • Semantic — purpose-named aliases (action/primary/default, text/primary, bg/surface). Components reference these.
  • Component-specific — optional per-component tokens when a component needs its own tuning knob (button/bg/primary). Use sparingly.

Sync flow

Figma Variables → figma-tokens.json → tokens/build.ts → tokens/dist/
                                                         ├── tokens.css
                                                         └── tokens.ts

Project structure

├── docs/           Documentation
├── tokens/         figma-tokens.json → CSS vars + TS
└── src/
    ├── components/ React components
    └── icons/      Figma → lucide-react catalog

2. Design principles

  1. Token-first — All visual values come from design tokens. No hardcoded colors, spacing, or typography in components.
  2. Semantic over raw — Semantic token names describe purpose, not appearance. Components reference destructive/default, not red/500.
  3. Figma as source of truth — Tokens and component specs live in Figma. Code is generated or guided by Figma exports.
  4. Mode-ready — All semantic tokens resolve through modes (Light/Dark, density, brand) so a single component definition adapts to any theme.

3. Token naming

Pattern

{category}/{purpose}/{state}
  • category — action, destructive, success, warning, info, text, bg, border, icon, overlay
  • purpose — primary, secondary, subtle, inverse, link, disabled
  • state — default, hover, pressed, focus, disabled, selected

Examples

Figma variableCSS custom property
action/primary/default--action-primary-default
action/primary/hover--action-primary-hover
destructive/default--destructive-default
text/primary--text-primary
bg/surface--bg-surface
border/default--border-default

Rules

  • Use lowercase, slashes in Figma, kebab-case as CSS vars.
  • Do not encode appearance in the name (red-hover is wrong; destructive-hover is right).
  • Stateful tokens always live at the third level (.../hover, .../pressed).
  • A token that exists on one category should exist on comparable categories (if action/primary/hover exists, so should destructive/hover).

4. Modes

Built-in modes

  • Theme: Light (default), Dark
  • Optional: Density (Comfortable, Compact), Brand (if multi-brand)

Rules

  • Every semantic token defines a value for every mode. No mode may be empty.
  • Components never branch on theme — they reference semantic tokens only.
  • Primitive collections do not need modes; modes live on the semantic layer.

Required dark-mode pairs (minimum)

SemanticLightDark
bg/canvas#FFFFFF#0B0D12
bg/surface#F7F8FA#141821
text/primary#0F1420#F2F4F8
text/secondary#4A5261#A7B0BF
border/default#E4E7EC#2A2F3A
action/primary/default#0097DB#3AB7F2

5. Component guidelines

File layout

src/components/[ComponentName]/
├── [ComponentName].tsx
├── [ComponentName].css
└── index.ts

CSS class naming (BEM-like)

  • Block: shain-[component] (e.g. shain-btn, shain-input)
  • Element: shain-[component]__[element] (e.g. shain-btn__label)
  • Modifier: shain-[component]--[modifier] (e.g. shain-btn--primary, shain-btn--sm)

Multiple modifiers combine: shain-btn shain-btn--primary shain-btn--md shain-btn--destructive.

Token usage in code

DoDon't
background-color: var(--action-primary-default)background-color: #0097DB
padding: var(--spacing-2) var(--spacing-3)padding: 8px 12px
border-radius: var(--radius-md)border-radius: 8px
color: var(--text-primary)color: #0F1420

Theme config pattern (variant styling)

For components with variants and states, use a config object instead of variant-specific CSS:

ts
enum ButtonVariant {
  Primary = "primary",
  Secondary = "secondary",
}

const theme = {
  [ButtonVariant.Primary]: {
    base:     ["--action-primary-default",  "--action-primary-default",  "--text-inverse"],
    hover:    ["--action-primary-hover",    "--action-primary-hover",    "--text-inverse"],
    pressed:  ["--action-primary-pressed",  "--action-primary-pressed",  "--text-inverse"],
    disabled: ["--action-disabled",         "--border-default",          "--text-disabled"],
  },
  [ButtonVariant.Secondary]: {
    base:     ["--bg-surface",              "--border-default",          "--action-primary-default"],
    hover:    ["--bg-surface-hover",        "--border-default",          "--action-primary-hover"],
    pressed:  ["--bg-surface-pressed",      "--border-default",          "--action-primary-pressed"],
    disabled: ["--bg-surface",              "--border-default",          "--text-disabled"],
  },
};
// Apply via CSS variables: --cmp-bg, --cmp-border, --cmp-color

Component anatomy

Interactive components support: variants (primary, secondary, outline, ghost, link), sizes (sm, md, lg, icon), states (default, hover, focus, pressed, disabled, loading).

Non-interactive components (e.g. Badge): no hover states or transitions, cursor: default.

Variant properties in Figma

Keep variant properties to 2–3 maximum. In exceptional cases 4 is acceptable with justification. Use boolean properties for optional elements (icons, badges) and instance swap for swappable elements.

Good — 9 variants + flexible toggles:
  Variant: Size (sm, md, lg) × Type (primary, secondary, outline)
  Boolean: Show Leading Icon, Show Trailing Icon
  Instance swap: Icon

Bad — 864 variants:
  Size (4) × Color (6) × Icon Position (3) × State (4) × Alignment (3)

Props convention

  • TypeScript interfaces with explicit types
  • Support className and spread remaining HTML attributes
  • forwardRef for form elements and focusable components
  • Default values match the most common use case

Icons

  • Use lucide-react
  • Accept LucideIcon type for icon props
  • Icon size: 16px inline with text; use tokens if a size scale exists
  • Catalog: src/icons/iconCatalog.ts maps Figma names → Lucide components

6. Accessibility

  1. Focus — :focus-visible for keyboard focus; avoid focus ring on mouse click.
  2. Disabled — disabled on interactive elements; aria-disabled only as fallback.
  3. Loading — aria-busy="true" when appropriate.
  4. Labels — All form controls have associated labels (visible or aria-label).
  5. Keyboard — Tab, Enter, Space, Escape for dropdowns/modals.
  6. Touch targets — Minimum 44×44px.
  7. Contrast — 4.5:1 for text, 3:1 for UI elements, verified in both Light and Dark.

7. Typography

  • Font family: sans-serif (configurable per brand)
  • Weights: 400 (normal), 600 (medium), 700 (bold)
  • Sizes: text/xs (12px) through text/3xl (32px)
  • Line heights: leading/1 (16px) through leading/6 (40px)

8. Spacing & Radius

Spacing scale:

TokenValue
spacing/00px
spacing/0.52px
spacing/14px
spacing/28px
spacing/312px
spacing/416px
spacing/520px
spacing/624px

Radius scale:

TokenValue
radius/none0px
radius/xs2px
radius/sm4px
radius/md8px
radius/lg12px
radius/xl24px
radius/full9999px

9. Anti-patterns

  1. Absolute positioning instead of auto layout — Always use auto layout with spacing tokens.
  2. Too many variants — 100+ variants means booleans and instance swaps are underused.
  3. Hardcoded values — Any raw hex, px value, or magic number should be a token reference.
  4. Appearance-based names — red-hover, blue-bg leak visuals into purpose names.
  5. Components referencing primitives — Primitives are internal; components reference semantic tokens.
  6. Unnamed layers — "Frame 123" is unacceptable. Name every layer by purpose.
  7. No component docs — Every component needs a doc from the template.
  8. Mixing presentation and content — Separate structural layers from content layers.
  9. Componentizing everything — Only reusable, consistent, interactive, or coded elements.
  10. Ignoring accessibility in Figma — 44×44px targets, contrast ratios, focus states.
  11. Deep nesting — Maximum 3 levels (e.g., Card → Button → Icon).
  12. Detached instances — Always use main component references inside other components.

10. Checklists

Figma-side (before handoff)

Structure: auto layout, descriptive layer names, ≤3 nesting levels, main component refs, proper constraints.

Properties: 2–3 variant props, booleans for optional elements, text props for content, instance swap for icons, sensible defaults.

Tokens: all fills use semantic variables (no hex), spacing variables, text styles, effect styles, no magic numbers. No primitive references from component layers.

States: default, hover, pressed/active, disabled, focus (keyboard).

Testing: works at different sizes, short/long content, no overflow, Light and Dark mode both pass contrast checks.

Code-side (new components)

  • All colors: var(--semantic-token-name) — never primitives
  • Variant colors via theme config
  • All spacing: var(--spacing-N)
  • All typography: var(--text-*) / var(--leading-*)
  • Border radius: var(--radius-*)
  • CSS classes: shain-[component] naming
  • TypeScript typed props
  • forwardRef for focusable elements
  • :focus-visible styles
  • Component doc from template

11. AI workflow for component generation

When generating components from Figma:

  1. Read component spec — Variants, sizes, states from Figma.
  2. Map to semantic tokens — Every fill, stroke, text, padding, radius, gap maps to a semantic token (never a primitive).
  3. Use the theme config pattern for variants with multiple states.
  4. Follow component guidelines — BEM naming (shain-*), forwardRef, accessibility.
  5. Verify both modes — Component must pass contrast in Light and Dark.
  6. Add changelog entry to the component doc.

Reference files

Read these when you need deeper context:

FileWhen to read
references/figma-best-practices.mdBuilding/reviewing Figma components, design tokens architecture, platform-specific guidelines (iOS/Android/Web), export settings
references/token-improvements.mdPlanning token system changes, adding new tokens, restructuring collections, brand expansion

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Shain Dls AI skill do?

Token-based design system skill for Figma + code handoff. Use whenever working with Figma components, design tokens, color variables, token migration, component creation, design-to-code handoff, or DS audits. Trigger on: design system, DS, DLS, tokens, figma variables, component spec, token naming, token hierarchy, which color to use, variant properties, theme config, dark mode. Even for "which color should I use" or "create a component" — use this skill.

Why use Shain Dls on TypingMind?

Because you install it once and use it with any model. Shain Dls 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 Shain Dls in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Dishain/Shain-Design-System-Skill/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Shain Dls?

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 Shain Dls?

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

Is the Shain Dls AI skill free?

Yes. It is published on GitHub by Dishain 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 👇