Design Tokens logo

Design Tokens

Community
TaewoooPark
design-tokens

UIForge's token foundry — emits and ENFORCES a project's design signature before any component is built. Use at the start of any UI work, or when asked to set up a theme, design system, design tokens, color palette, type scale, or spacing/radius/shadow foundation, or to make a shadcn/Tailwind theme look non-default. Emits tokens.css (color roles, type scale, 8px spacing, radius, shadow) + motion.ts derived from the chosen design direction, then scans the build for off-token "magic numbers" and corrects them — so every value in the UI traces to one coherent source, the way v0 and Lovable produce premium output.

Overview

PublisherTaewoooPark
RepositoryUIForge
Skill namedesign-tokens
Stars
8
Forks
1
Bundled files
3
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.

  • 3 bundled files

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

  • Open source

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

Installation

Install the Design Tokens 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/TaewoooPark/UIForge/tree/main/skills/design-tokens
  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/TaewoooPark/UIForge.git /tmp/UIForge
mkdir -p .claude/skills
cp -r /tmp/UIForge/skills/design-tokens .claude/skills/design-tokens
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Design Tokens 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 Design Tokens 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 Design Tokens 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.

Design Tokens — emit first, enforce always

Token-first, enforced at generation time, is the single mechanism most correlated with "designed vs slop" output across every premium AI-UI tool (v0 mandates tokens; Lovable scans each generation for violations and auto-corrects; Subframe cascades them). Do the same: emit the signature before the first component, then let nothing use a value that isn't in it.

1. Emit tokens.css before any component

Don't start from a blank page — that's how the median wins. Start from a kit: copy the chosen direction's ready theme from ${CLAUDE_PLUGIN_ROOT}/tools/kits/ (each ships a real, non-default typeface, a committed palette, radius/shadow, and a reduced-motion path — and already passes the linter). Install its fonts (the header line: npm i @fontsource…). Or run /uiforge:reskin <image|url> to derive tokens from a reference. The bare ${CLAUDE_PLUGIN_ROOT}/tools/tokens.template.css is the fallback when no kit fits. Tune the roles to the brand, but keep components consuming var(--…) or the generated utilities (bg-bg, text-fg, bg-primary, bg-surface, text-muted, border-border) — never literals:

css
:root {
  /* color — semantic ROLES, not raw hex at point of use */
  --bg: …; --fg: …;                 /* base surface + text */
  --surface: …; --on-surface: …;    /* raised card + its text */
  --muted: …; --border: …;          /* secondary text, hairlines */
  --primary: …; --on-primary: …;    /* the ONE accent + its text */
  --ring: …;                        /* focus ring, ≥3:1 */

  /* space — 8px grid (4px half-step); ONE scale */
  --space-1: 4px; --space-2: 8px; --space-3: 12px; --space-4: 16px;
  --space-6: 24px; --space-8: 32px; --space-12: 48px; --space-16: 64px;

  /* radius — one vocabulary */
  --radius-sm: …; --radius: …; --radius-lg: …;

  /* elevation — one light source, monotonic */
  --shadow-1: 0 1px 2px rgb(0 0 0 / .06);
  --shadow-2: 0 2px 6px rgb(0 0 0 / .08);
  --shadow-3: 0 8px 24px rgb(0 0 0 / .12);

  /* type — sizes from ONE ratio; role line-heights */
  --text-xs: …; --text-sm: …; --text-base: 16px; --text-lg: …;
  --text-xl: …; --text-2xl: …; --text-3xl: …; --text-display: …;
  --leading-tight: 1.1; --leading-snug: 1.25; --leading-normal: 1.5;
  --font-display: …; --font-text: …; --font-mono: …;
}

Fill each … from the three references, tuned to the direction. Add a .dark / @media (prefers-color-scheme: dark) block that swaps the same roles.

Motion tokens (motion.ts — easing/duration/spring) belong to the same signature; the motion skill owns their exact values. Emit it in the same first step.

2. Override the defaults (this is a real rule)

Unmodified shadcn slate/zinc + the stock --radius + a default font is the #1 data-mined tell of AI-built UI. Change all three and add one token the default kit doesn't ship (a signature surface, an ink accent, a distinctive radius) and apply it consistently. tweakcn can generate a non-default shadcn theme to start from.

3. Enforce — run the linter (the Gate)

Token discipline is not a vibe; it's a gate. Run the real linter — it exits non-zero on off-token values and slop tells:

bash
node ${CLAUDE_PLUGIN_ROOT}/tools/uiforge-lint.mjs <project-dir> --strict

Fix every BLOCKER (default font, AI purple, gradient headline, emoji-as-UI, hype copy, no reduced-motion) and resolve the warnings (raw hex, arbitrary values, off-grid spacing). Re-run until it exits 0. For reference, the checks mirror these patterns (promote each hit to a token or snap it to the scale):

# raw hex / rgb at point of use (should be var(--…))
#[0-9a-fA-F]{3,8}\b   |   rgb\(|rgba\(|hsl\(
# off-grid spacing (not a multiple of 4) in p-/m-/gap-/space- or px values
\b(p|m|gap|space)[trblxy]?-\[?\d*\.?\d+px\]?     # inspect: on the 4/8 grid?
[^-]\b(5|7|9|11|13|14|15|17|18|19|21|22|23)px\b  # common off-scale px
# arbitrary type sizes / line-heights bypassing the scale
text-\[\d+px\]   |   leading-\[\d

Every hit is either promoted to a token or snapped to the nearest scale step. "No component invents its own visual properties; it selects from the tokens."

4. Hand back

Report the emitted tokens (the palette, the ratio + resulting sizes, the spacing scale, radius, shadows, fonts) so the director and the other skills build on one source of truth. If you changed a default, say which and why.

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 Design Tokens AI skill do?

UIForge's token foundry — emits and ENFORCES a project's design signature before any component is built. Use at the start of any UI work, or when asked to set up a theme, design system, design tokens, color palette, type scale, or spacing/radius/shadow foundation, or to make a shadcn/Tailwind theme look non-default. Emits tokens.css (color roles, type scale, 8px spacing, radius, shadow) + motion.ts derived from the chosen design direction, then scans the build for off-token "magic numbers" and corrects them — so every value in the UI traces to one coherent source, the way v0 and Lovable pro...

Why use Design Tokens on TypingMind?

Because you install it once and use it with any model. Design Tokens 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 Design Tokens in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/TaewoooPark/UIForge/tree/main/skills/design-tokens. 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 Design Tokens?

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 Design Tokens?

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

Is the Design Tokens AI skill free?

Yes. It is published on GitHub by TaewoooPark 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 👇