Taste Design System logo

Taste Design System

Community
madebymustafa
taste-design-system

Builds and audits design systems for agentic workflows: three-tier design tokens (primitive → semantic → component), W3C DTCG token format, spacing and type scales, color ramps, component anatomy, documentation, governance, and adoption metrics. Use when creating tokens, scaffolding a component library, or auditing an existing system for consistency. Stack on the design-taste core skill.

Overview

Publishermadebymustafa
Repositorydesign-taste
Skill nametaste-design-system
Stars
198
Forks
10
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 madebymustafa on GitHub. Read the source before you install it.

Installation

Install the Taste Design System 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/madebymustafa/design-taste/tree/main/skills/taste-design-system
  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/madebymustafa/design-taste.git /tmp/design-taste
mkdir -p .claude/skills
cp -r /tmp/design-taste/skills/taste-design-system .claude/skills/taste-design-system
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Taste Design System 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 Taste Design System 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 Taste Design System 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.

taste-design-system

A design system is a decision inventory. This skill tells the agent how to take that inventory, structure it, and keep it honest.

1. TOKEN ARCHITECTURE (Three Tiers, Never Two)

  1. Primitive tokens — raw values: --color-ink-900, --space-4, --radius-md. No meaning beyond the value.
  2. Semantic tokens — meaning, not value: --color-surface, --color-text-muted, --color-border, --space-inset-lg. Products consume semantic tokens ONLY.
  3. Component tokens — per-component aliases: --button-bg-primary, --button-radius, --input-border-width. Components never reference primitives directly.

Rule: components → semantic → primitive → value. If any layer is skipped, the system will fork.

2. TOKEN FORMAT (DTCG, Not Freeform)

Write tokens as a flat key map (or W3C DTCG JSON if the stack supports it):

css
--color-ink-900: #101014;
--color-ink-700: #3B3B40;
--color-accent: #E8540B;
--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-sm: 2px; --radius-md: 6px; --radius-lg: 12px;
--font-display: 'Archivo', sans-serif; --font-body: 'Inter', sans-serif; --font-mono: 'JetBrains Mono', monospace;

Naming conventions: kebab-case, noun-first, scale-last (--space-4, --color-ink-900, --radius-md). Never encode intent in names ("fancy-purple"), never use hex colors directly in components.

3. THE INVENTORY (Audit Before You Build)

Run the inventory before building anything:

  1. Colors — list every hex in the codebase. Flag any value not in the token set (>15 unique out-of-scheme colors = system is broken).
  2. Spacing — flag any margin/padding not on the 4px (or 8px) scale.
  3. Type — count distinct font sizes/weights in use. >5 sizes on one page = no hierarchy.
  4. Radius / elevation / borders — collect every value; cluster into 2–4 tokens.
  5. Components — list every component; flag components with >3 variants doing the same job, and any component styling itself from raw values.

Output: a token map, a violation list with counts, and the fix order (primitives first, then semantic mapping, then component cleanup). Never refactor a system in the same pass as auditing it — audit, report, then change.

4. SCALES (Fixed Reference)

  • Spacing: 4/8/12/16/24/32/48/64/96. Gaps below 4px only for hairline borders. Never 5, 7, 10, 14px — unless a deliberate optical override is documented.
  • Type scale (fluid): 12/14/16/18/24/32/48/64 rendered with clamp() where responsive.
  • Color ramps: 9–11 steps from 50→950 in OKLCH (perceptually even), neutrals and at least one accent family. Dark mode = semantic mapping change, not new colors.
  • Elevation: 3 levels max: flat (no shadow), raised (subtle, blur ≥ 16px, opacity ≤ 0.15), floating (blur ≥ 40px, opacity ≤ 0.2). Everything else is a bug.

5. COMPONENT ANATOMY (Every Component Ships With)

  1. One job, one name. If a component has an "and" in its description, split it.
  2. Full state matrix: default, hover, active/pressed, focus-visible, disabled, loading, error. If a state is impossible, document why.
  3. Size variants only where real (sm/md/lg), never color variants pretending to be states.
  4. Responsive behavior and the breakpoint where it changes.
  5. Token references in code — zero raw values inside component styles.

6. DOCUMENTATION & GOVERNANCE

  • One page per component: purpose, anatomy diagram (as code comment), state matrix, usage rules, do/don't examples, accessibility notes.
  • Change process: token changes are additive (new token → deprecate old → remove after one release cycle). Never mutate a token value silently.
  • Ownership: one person (or one skill) holds the token file. Component decisions route through it.
  • Adoption metrics (measure quarterly): % of screens consuming semantic tokens, # of raw-color violations, # of off-scale spacing values, % of components with documented states. The number that matters: violation count trending down.

7. CHECKLIST (Fail the System If…)

  • Any component references a primitive directly.
  • Any UI uses a hex not present in the token set.
  • Any spacing value off the scale without a documented override.
  • Any component missing a disabled or focus-visible state.
  • More than 3 elevation levels in use.
  • Token names encode meaning AND value ("dark-purple" instead of semantic).
  • Two components exist that do the same job with different APIs.

Tokens are the memory of a design system. If the system can't remember, it repeats itself — and so do your pages.

Frequently asked questions

What does the Taste Design System AI skill do?

Builds and audits design systems for agentic workflows: three-tier design tokens (primitive → semantic → component), W3C DTCG token format, spacing and type scales, color ramps, component anatomy, documentation, governance, and adoption metrics. Use when creating tokens, scaffolding a component library, or auditing an existing system for consistency. Stack on the design-taste core skill.

Why use Taste Design System on TypingMind?

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

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

Which AI models can use Taste Design System?

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 Taste Design System?

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

Is the Taste Design System AI skill free?

Yes. It is published on GitHub by madebymustafa 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 👇