Taste Typography logo

Taste Typography

Community
madebymustafa
taste-typography

Typography craft for agent-built interfaces: pairing rules, fluid type scales, line lengths, baseline rhythm, weight discipline, legibility minimums, and the typography reasons most AI-generated pages read as cheap. Use whenever generating or reviewing any UI with text. Stack on the design-taste core skill.

Overview

Publishermadebymustafa
Repositorydesign-taste
Skill nametaste-typography
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 Typography 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-typography
  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-typography .claude/skills/taste-typography
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Taste Typography 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 Typography 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 Typography 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-typography

Type is the first thing people register and the last thing they can name. This skill makes sure the type does the design's work.

1. PAIRING (Two Families Max, With Purpose)

Pick two families with structural contrast — never two similar ones:

DisplayBodyWorks for
Serif with character (Fraunces, Newsreader)Neutral sans or the serif itselfEditorial, luxury, calm
Heavy grotesk (Archivo, Bricolage)Same family at 400Bold, playful, brutalist
Neutral grotesk (Space Grotesk, system)Same familyDark tech, developer tools
Mono (labels/metadata only)Grotesk bodyTechnical, terminal
  • Three families = soup. One neutral sans = template.
  • Mono is a label voice, not a body voice. Mono for body text is for data tables and terminals only.
  • If the brand mandates a family, the pairing is built around it — the second family comes from the same species (serif + sans from the same designer is a safe pair).

2. THE SCALE (Two Scales, Applied Consistently)

Display scale (marketing, heroes): 64 / 48 / 32 / 24 / 18 / 16 — apply with clamp() for fluid size: clamp(2rem, 5vw, 4rem).

UI scale (products, dashboards): 13 / 14 / 15 / 16 / 20 / 24 — never below 12px, never above 24px inside UI surfaces.

Hierarchy rules:

  • One display size per section. If two headings on a page are equally large, neither is a heading.
  • Weights: one weight for display (the family's natural voice: 400 for serifs, 700–900 for grotesks), 400 for body, 500–600 for emphasis. Three weights per page max.
  • Small type (≤ 12px) must be accompanied by letter-spacing ≥ 0.02em or it renders muddy — but never below 10px without a data-table justification.

3. MEASURE, LEADING, RHYTHM (The Numbers)

  • Line length: 45–75 characters for body. Hero subheads ≤ 40 chars per line, headlines ≤ 24 chars per line (break early, not by luck).
  • Leading: headlines 0.95–1.08 (serifs need the low end), body 1.55–1.7, labels 1.2–1.4.
  • Letterspacing: uppercase labels 0.08–0.2em; display lowercase −0.02em; body 0. NEVER apply tracking to lowercase body text above 16px except as a deliberate editorial act.
  • Vertical rhythm: all text blocks align to the 4px baseline grid. When in doubt, line-height in multiples of 4: 20/24/28/32px.
  • Paragraphs: max ~3 lines per paragraph in marketing copy. A 6-line paragraph in a card is a layout failure, not a copy failure.

4. TEXT PLACEMENT (The Details That Read as "Designed")

  • Never set body text full-width — constrain to 32–40em columns, even (especially) on widescreen.
  • Break headlines deliberately: control word breaks with manual line structure where it matters ("Know exactly / what your users do").
  • Headline + sub + paragraph: each step should visibly shrink in size AND weight, not just one of them.
  • Justified text is banned except for print-style editorial columns ≥ 60 chars with hyphenation. Left-aligned everywhere else.
  • Ragged edges matter: no widows (single word on last line) in headings — fix with word breaks or sizing.
  • Faux styling: no fake italics (use the family's real italic face), no fake bold (never font-weight: 700 on a 400-only family).
  • Fluid type: clamp() with rem, not px, for all text. Never vw units without a rem floor.

5. ACCESSIBILITY FLOOR (Non-Negotiable)

  • Body ≥ 16px on marketing pages, ≥ 14px inside UI; below that, only metadata with 4.5:1 contrast.
  • Contrast ≥ 4.5:1 for all body, 3:1 for large display (≥ 24px / 18.66px bold).
  • text-rendering: optimizeLegibility on display; no -webkit-text-stroke (breaks at small sizes).
  • Font loading: font-display: swap; preload the display face; never render invisible text waiting for fonts.

6. CHECKLIST (Fail the Page If…)

  • One neutral family doing every job.
  • More than two families.
  • A heading below 20px with body beside it at 16px (no hierarchy).
  • Body text wider than 75 characters.
  • Any text under 10px.
  • More than three font weights in use.
  • A 6-line paragraph inside a card or sidebar.
  • Justified text without hyphenation support.

If the typography works, people will say the site "feels premium" and not know why. That's the goal.

Frequently asked questions

What does the Taste Typography AI skill do?

Typography craft for agent-built interfaces: pairing rules, fluid type scales, line lengths, baseline rhythm, weight discipline, legibility minimums, and the typography reasons most AI-generated pages read as cheap. Use whenever generating or reviewing any UI with text. Stack on the design-taste core skill.

Why use Taste Typography on TypingMind?

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

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

Which AI models can use Taste Typography?

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 Typography?

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

Is the Taste Typography 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 👇