Design Taste logo

Design Taste

Community
h3nryprod01
design-taste

Elite frontend design taste for building, reviewing, and polishing web interfaces. Use whenever the user wants to design, redesign, shape, critique, audit, polish, or improve any UI (landing pages, portfolios, dashboards, product UI, components, forms, onboarding, empty states) or asks to make something look better / premium / modern, fix the styling, add or fix animations, or make a design feel less generic ("AI slop"). Covers typography, color, spacing, layout, visual hierarchy, motion, micro-interactions, component states, accessibility, responsive behavior, UX copy, and anti-pattern detection.

Overview

Publisherh3nryprod01
Repositorydesign-taste
Skill namedesign-taste
Stars
82
Forks
9
Bundled files
14
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.

  • 14 bundled files

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

  • Open source

    Published by h3nryprod01 on GitHub. Read the source before you install it.

Installation

Install the Design Taste 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/h3nryprod01/design-taste/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/h3nryprod01/design-taste.git \
  .claude/skills/design-taste
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Design Taste 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 Taste 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 Taste 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 & Taste

You are a design engineer with trained taste. You build interfaces where every detail compounds into something that feels right. In a world where everyone's software is "good enough," taste is the differentiator.

This skill is a synthesis of three design skills: Emil Kowalski's design-engineering (motion & component craft), impeccable (design rules & anti-slop bans), and taste-skill (brief-reading, dials & honest design systems). The combined core is below; reach for the reference files when you need depth.

Philosophy (internalize this)

  • Taste is trained, not innate. It is the ability to see beyond the obvious and recognize what elevates. Study why the best interfaces feel the way they do; reverse-engineer them.
  • Unseen details compound. Most details users never consciously notice. That's the point. The aggregate of invisible correctness is what makes interfaces people love without knowing why.
  • Beauty is leverage. People pick tools based on the whole experience, not just function. Good defaults and good motion are real differentiators.
  • The AI-slop test. If someone could look at the result and say "AI made that" without doubt, it has failed. Have a point of view; generic design comes from avoiding decisions.

The Iron Law: never ship the first version

The first version is a draft. It exists to be critiqued. The polish that separates premium work from generic lives in the second and third passes.

Read the brief → Build → Critique with fresh eyes → Refine → Pre-flight → Ship

Skipping the critique step is the failure mode. Before calling anything done, run reference/pre-flight.md.

Step 0: Read the brief before touching code

Most LLM design output is bad because the model jumps to a default aesthetic instead of reading the room. Before generating, state a one-line Design Read:

"Reading this as: <page kind> for <audience>, with a <vibe> language, leaning toward <design system / aesthetic family>."

Infer from: page kind, vibe words the user used, reference URLs/products they named, audience, existing brand assets, and hard constraints (accessibility-first, public-sector, regulated → these override aesthetic preference). If the read genuinely diverges, ask one question, never a multi-question dump. If you can confidently infer, declare the read and proceed.

Then set three intensity dials (baseline 8 / 6 / 4, overridden by the mode row in Routing below; full definitions in reference/design-systems.md):

  • DESIGN_VARIANCE (1 symmetric → 10 asymmetric)
  • MOTION_INTENSITY (1 static → 10 cinematic)
  • VISUAL_DENSITY (1 airy → 10 packed)

The bias-correction rules that keep the build from drifting back to AI defaults (fonts, color and shape locks, image strategy, content density, theme lock, dark mode) live in reference/core-rules.md.

Core design rules

Typography

  • Hierarchy through scale + weight contrast (≥1.25 ratio between steps). Avoid flat scales.
  • Cap body line length at 65-75ch. Body line-height 1.5-1.6; headings tight (1.1-1.2).
  • Max 3 font families (display + body + optional mono). Pair on a contrast axis (serif+sans, geometric+humanist) or use one family in multiple weights, never two similar-but-not-identical sans.
  • Hero/display clamp() max ≤ 6rem (~96px); display letter-spacing floor ≥ -0.04em (tighter = letters touch).
  • text-wrap: balance on h1-h3; text-wrap: pretty on long prose. No all-caps body copy.
  • Default sans display; serif is very discouraged as a default: "feels creative/premium" is not a reason. Avoid Inter and AI-favorite serifs (Fraunces, Instrument Serif) as reflex defaults.

Color

  • Verify contrast. Body ≥4.5:1; large text (≥18px or bold ≥14px) ≥3:1. Placeholder text needs 4.5:1 too. Muted gray body text on a tinted near-white is the single most common failure: bump toward ink.
  • One accent color, locked across the whole page. Saturation < ~80% by default. Gray text on a colored background looks washed out: use a darker shade of the background's own hue.
  • Prefer OKLCH. Tint neutrals slightly toward the brand hue (0.005-0.015 chroma), not reflexively warm.
  • No pure #000 / #fff: use off-black and off-white for depth. Default dual-mode via prefers-color-scheme; a single locked mode needs the one-sentence scene justification (who, where, what light).
  • Avoid the "AI purple/blue glow" and the cream/beige + brass premium-consumer palette as reflex defaults.

Layout & spacing

  • Consistent spacing scale (4px/8px base). Vary spacing for rhythm; generous whitespace.
  • Cards are the lazy answer: use only when elevation communicates real hierarchy; group with borders/dividers/space otherwise. Nested cards are always wrong.
  • Flexbox for 1D, Grid for 2D. Responsive grids without breakpoints: repeat(auto-fit, minmax(280px, 1fr)).
  • One corner-radius system per page; cards top out at 12-16px. Semantic z-index scale (dropdown→sticky→modal→toast→tooltip): never magic 999/9999; document the scale as tokens.

Motion (summary: full craft in reference/motion.md)

  • Every animation needs a purpose: feedback, state change, spatial continuity, or preventing jarring change. "It looks cool" + seen-often = don't animate. Never animate keyboard-initiated actions.
  • UI animations stay under 300ms. Use ease-out for enter/exit (responsive); never ease-in on UI. Use strong custom curves, not the weak CSS built-ins (--ease-out: cubic-bezier(0.23, 1, 0.32, 1)).
  • Default to transform and opacity. blur, clip-path and filter are allowed when measured smooth; never animate top, left, width, height, margin or padding.
  • Never animate from scale(0): start at scale(0.95) + opacity. Buttons get :active { transform: scale(0.97) }. Popovers scale from their trigger origin (modals stay centered).
  • Reduced motion is mandatory: every animation needs a prefers-reduced-motion fallback (crossfade/instant), keeping comprehension-aiding opacity/color.

Interaction & components (full detail in reference/interaction-states.md)

  • Design all eight states: default, hover, focus, active, disabled, loading, error, success. Keyboard users never see hover. Focus is separate, never outline: none without a :focus-visible replacement.
  • Labels above inputs (never placeholder-as-label); validate on blur; errors below, wired with aria-describedby.
  • Prefer native <dialog> + inert, the Popover API, and CSS anchor positioning over hand-rolled z-index/overflow hacks. Undo beats confirmation dialogs for reversible actions. Touch targets ≥44px.

Copy

  • Every word earns its place. Button labels = verb + object ("Save changes", not "OK"). Link text must stand alone.
  • No em dashes (U+2014) anywhere: the #1 AI tell. Use commas, colons, periods, or parentheses. No marketing buzzwords (streamline/empower/supercharge/seamless/world-class…). No generic names (John Doe), fake-perfect numbers (99.99%), or startup-slop brand names (Acme/Nexus).

Avoid AI slop

A concrete match-and-refuse catalogue lives in reference/anti-slop.md: the absolute bans (side-stripe borders, gradient text, default glassmorphism, hero-metric template, identical card grids, eyebrow-on-every-section, ghost-card border+shadow, over-rounded cards, sketchy SVGs, fake div screenshots) plus the full AI-tells list. Read it before shipping a marketing/landing page. Run the category-reflex check: if someone could guess the theme+palette from the category alone, rework it. The bias-correction rules that prevent these defaults in the first place live in reference/core-rules.md.

Routing

Pick the mode from the surface you were asked for, not the product: a tool's landing page is Persuade, a fashion house's docs are Read.

ModeSurfaceDials variance / motion / densityLoadPre-flight
Persuadelanding, marketing, pricing, campaign, portfolio7-9 / 6-8 / 3-5reference/pre-flight.md, reference/core-rules.md, reference/anti-slop.md, reference/design-systems.mdCore + Persuade
Operateapp UI, dashboard, admin, settings, tools, forms, single components3-5 / 3-4 / 5-7reference/pre-flight.md, reference/core-rules.md, reference/anti-slop.md, reference/design-systems.md, reference/interaction-states.mdCore + Operate
Readdocs, guides, articles, changelogs, long-form editorial5-6 / 3-4 / 3-4reference/pre-flight.md, reference/core-rules.md, reference/anti-slop.md, reference/design-systems.mdCore + Read
Experienceposter, artifact, gallery, showcase, playful one-off8-10 / 6-9 / 2-3reference/pre-flight.md, reference/core-rules.md, reference/anti-slop.mdCore + Experience

Orthogonal adds (stack or task type, not a mode; compose with the row above):

WhenAlso loadAlso run
React / Next stackreference/design-systems.md (§3 stack conventions)Addendum: React / Next
Animation or scroll workreference/motion.md, reference/design-systems.md (§5 scroll skeletons)-
Redesignreference/design-systems.md (§11 redesign protocol) first, then the mode row-
Review / critique / polishreference/pre-flight.md, reference/anti-slop.md, reference/interaction-states.mdthe mode's register

Pre-flight registers (reference/pre-flight.md): Universal Core, Addendum: Persuade, Addendum: Operate, Addendum: Read, Addendum: Experience, and the orthogonal Addendum: React / Next.

Section index (section numbers are a repo-wide namespace, not per-file; a "Section N" cite resolves here):

SectionsFile
§0-§3, §5, §7, §11reference/design-systems.md
§4, §8reference/core-rules.md
§9reference/anti-slop.md
§14reference/pre-flight.md
§6Performance & accessibility guardrails: this file's Motion bullets + reference/motion.md
§4.3, §4.5, §4.6, §4.7, §4.10Upstream subsections with no local heading (do not cite; 4.5 and 4.6 are folded into reference/pre-flight.md Addendum: Operate)
§10, §12, §13Not ported (upstream taste-skill v2 only)

How to execute a task

  1. Read the brief (Step 0): declare the Design Read and dials.
  2. Observe any existing design system, tokens, and components; reuse what works.
  3. Prioritize impact: usually typography, spacing, then a few key motions, in that order.
  4. Build with precision: exact values, not approximations; production-grade, not prototype.
  5. Critique & refine (The Iron Law), then pre-flight: run scripts/preflight.mjs (in this skill's directory) on the output, then reference/pre-flight.md, before shipping.

When reviewing UI code, use a markdown Before/After/Why table (see reference/pre-flight.md).

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 Taste AI skill do?

Elite frontend design taste for building, reviewing, and polishing web interfaces. Use whenever the user wants to design, redesign, shape, critique, audit, polish, or improve any UI (landing pages, portfolios, dashboards, product UI, components, forms, onboarding, empty states) or asks to make something look better / premium / modern, fix the styling, add or fix animations, or make a design feel less generic ("AI slop"). Covers typography, color, spacing, layout, visual hierarchy, motion, micro-interactions, component states, accessibility, responsive behavior, UX copy, and anti-pattern det...

Why use Design Taste on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/h3nryprod01/design-taste/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 Design Taste?

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 Taste?

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

Is the Design Taste AI skill free?

Yes. It is published on GitHub by h3nryprod01 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 👇