Taste Code Craft logo

Taste Code Craft

Community
madebymustafa
taste-code-craft

Code-level craft for agent-built UI: semantic HTML, token-driven CSS (no hardcoded values), specificity discipline, component hygiene, zero inline styles, no dead code. The engineering half of "designed, not generated" — a beautiful page is betrayed by sloppy markup. Use when generating or auditing any frontend code. Stack on the design-taste core skill.

Overview

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

Use it in TypingMind

Enable Taste Code Craft 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 Code Craft 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 Code Craft 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-code-craft

A designed page is a designed codebase. The DOM is the first layout you write — structure it like one.

1. SEMANTIC HTML (The DOM Is Design)

  • One <h1> per page; heading levels descend without skipping (h1 → h2 → h3, never h2 after a bare h4).
  • Use the element that means the thing: <nav>, <main>, <section aria-labelledby>, <figure>, <time>, <button> (not a <div> with a click handler), <a href> for navigation.
  • Lists are <ul>/<ol> — a row of features is a list, not five <div>s.
  • Every <img> has alt (empty alt="" only when decorative) and explicit width/height to prevent layout shift.
  • Labels: every <input> has a visible <label> (or aria-labelledby); never placeholder as the only label.
  • If the page has no <main>, one <section> without a heading, or a clickable <div>, it fails.

2. TOKENS OVER LITERALS (The Design System in Code)

  • Every color, spacing value, radius, shadow, and font size comes from a variable — zero hardcoded hex values in component rules. Literal #000 or padding: 13px appearing three times is a token missing.
  • The palette in code matches the design read exactly: no invented extra grays (#F7F8FA for a page that says #F4F1EA), no brand-name hexes that weren't chosen.
  • Type uses the declared scale (rem/fluid clamp), not ad-hoc pixel sizes per component.
  • Breakpoints come from a single media-query convention (one set of widths, documented once), not per-component magic numbers.
  • When a design system exists (taste-design-system), components consume semantic tokens — never reach past them to primitives.

3. SPECIFICITY & CSS ARCHITECTURE

  • No !important — if one is needed, the selector model is wrong.
  • No styling by ID, no bare element selectors with side effects (section { }), no * { } resets past a single box-sizing reset.
  • Max nesting depth 2 (or the framework's own convention); selectors that need 4+ levels describe a component boundary that should exist.
  • Utility/inline systems (Tailwind) are fine if the config is tokenized and class counts stay readable; 30-class soup on one element is not.
  • Dead rules: any CSS rule that matches nothing in the markup gets deleted before delivery — pages ship with zero unused selectors.
  • Components are single-responsibility: one file/block per UI component, no component doing hero + footer + modal work.

4. ZERO INLINE STYLES (Except The Three Exceptions)

  • Inline style="..." is banned for layout, color, and type — that's what classes exist for.
  • Allowed exceptions (use sparingly, comment why): dynamic values (CSS variable injection from JS state, e.g. progress), one-off alignment of a single static element that would need a new class, SVG presentation attributes.
  • An audit counts inline styles; more than ~5 on a page = architecture drift.

5. HYGIENE & DELIVERY

  • No dead markup: unused wrapper divs, empty elements, comments of old code, console.log, TODO-left-in-production.
  • No placeholder assets: no lorem text, no https://placehold.co, no unsplash URLs without intent, no image that 404s.
  • Files that ship: every font is loaded once (see taste-web-performance), every stylesheet is used, no duplicate font links from copy-paste.
  • Formatting is consistent (one quote style, one trailing-comma policy) — run the formatter; an agent that skips formatting signals it skipped other things.
  • No accessibility regressions introduced in markup: focusable elements reachable, aria-hidden not applied to focusable content, buttons not nested in buttons.

6. CHECKLIST (Fail the Code If…)

  • Any element that should be a <button>/<a>/<ul> is a <div>.
  • Any hardcoded color, spacing, radius, or type size outside the token set.
  • Any !important, ID-styling, or 4+ level nesting.
  • Any inline style outside the three exceptions.
  • Any unused CSS rule, dead markup, or console.log shipped.
  • Heading levels skipped or duplicated h1.
  • An image without alt and explicit dimensions.
  • The code looks like it was generated by an agent that didn't read the design system.

Frequently asked questions

What does the Taste Code Craft AI skill do?

Code-level craft for agent-built UI: semantic HTML, token-driven CSS (no hardcoded values), specificity discipline, component hygiene, zero inline styles, no dead code. The engineering half of "designed, not generated" — a beautiful page is betrayed by sloppy markup. Use when generating or auditing any frontend code. Stack on the design-taste core skill.

Why use Taste Code Craft on TypingMind?

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

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

Which AI models can use Taste Code Craft?

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 Code Craft?

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

Is the Taste Code Craft 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 👇