Frontend Designer Lite logo

Frontend Designer Lite

Community
kozz36
frontend-designer-lite

Trigger: frontend design, CSS architecture, tokens, responsive UI, accessibility, motion. Produce production UI design-system decisions without local references.

Overview

Publisherkozz36
Repositoryfrontend-designer-skill
Skill namefrontend-designer-lite
Stars
8
Forks
1
Bundled files
Instructions only
LicenseApache-2.0
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 kozz36 on GitHub. Read the source before you install it.

Installation

Install the Frontend Designer Lite 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/kozz36/frontend-designer-skill/tree/main/skills/frontend-designer-lite
  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/kozz36/frontend-designer-skill.git /tmp/frontend-designer-skill
mkdir -p .claude/skills
cp -r /tmp/frontend-designer-skill/skills/frontend-designer-lite .claude/skills/frontend-designer-lite
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Frontend Designer Lite 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 Frontend Designer Lite 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 Frontend Designer Lite 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.

Activation Contract

Use this standalone skill for frontend design-system decisions when an agent must choose, review, or document production-ready technical direction.

  • Designing or reviewing UI systems, component anatomy, tokens, layout, or visual interaction.
  • Choosing CSS architecture, responsive behavior, accessibility constraints, or motion rules.
  • Converting repository visual direction into implementation-ready frontend guidance.

Do not use it for generic explanation, copy editing, or one-off code changes that do not affect architecture or reusable implementation patterns.

Non-Negotiable Runtime Rules

  • Preserve repository token and component-state contracts; do not hardcode isolated visual decisions.
  • Use semantic HTML and native controls before ARIA substitutes. Define native keyboard behavior, visible focus, measurable contrast, touch/pointer criteria, and reduced-motion behavior for every interactive UI.
  • Keep components reusable and responsive; reserve media and font geometry and do not accept performance regressions without compatible evidence.
  • Keep the answer decision-first. Do not add version, browser, API, accessibility-law, security, adoption, or legal-applicability claims without current authoritative verification.
  • Treat security requirements as conditional product and repository inputs; do not infer controls, policy, telemetry, or external-tool permission when evidence is absent.

Required Evidence Gates

Gather the following before selecting a pattern. If a required input is unavailable, stop and report the missing evidence rather than guessing or prescribing a universal stack.

  • Repository architecture and migration: inspect existing CSS architecture, naming, linting, build pipeline, team conventions, and accepted migration cost. Preserve the established paradigm; introduce cascade layers, CSS Modules, CSS-in-JS, utilities, or naming conventions only with repository support and an accepted migration plan.
  • Design source: inspect Design.md, design.md, token sources, a design-system package, design-tool export, or another approved source when present. Treat its case-sensitive path and token, semantic, component, and state contracts as local visual truth; reconcile conflicts with code and team conventions. Do not require a filename, rename a source, or assume a design tool.
  • Browser and fallback policy: inspect the supported browser matrix before using CSS nesting, container queries, :has(), light-dark(), scroll-driven animation, WAAPI, observers, or related APIs. Define a compatible fallback when required.
  • Tokens, color, and themes: preserve primitive, semantic, and component tiers or the local equivalent, including loading, disabled, error, success, and selected states. Use OKLCH-only for newly specified tokens only when the approved token contract and browser/build matrix select it; otherwise retain the compatible format. Measure every foreground/background and focus pair. Use dark-mode-first only when product requirements and the approved design source require it; otherwise preserve the existing semantic-token and theme mechanism.
  • Layout and components: select component-local container queries only when supported and valuable; use viewport queries for document-level layout, navigation, or the established responsive system. Choose mobile-first, desktop-first, grid, flex, named areas, or composition from information architecture, content, component ownership, and viewport evidence. Use semantic structure: accessible names for regions when needed, native <dialog> when it fits policy, native table semantics for tabular data, explicit form labels/errors/order/states, and <nav> with list semantics where applicable. Reserve media geometry with aspect-ratio, dimensions, or an equivalent contract when it reduces measured layout shift.
  • Accessibility and motion: preserve semantic HTML, labels, names, roles, states, keyboard order, persistent visible :focus-visible, unobscured focus, measurable contrast, and applicable touch-target criteria. Add ARIA only when native semantics cannot express the contract. Honor prefers-reduced-motion with equivalent low-motion or static feedback. Use the smallest motion mechanism and gate WAAPI, scroll-driven CSS, observers, and listeners on browser support, motion policy, and measured main-thread cost. Apply jurisdiction-specific obligations only when product scope establishes them.
  • Visual treatment and performance: use bento-like grids, glass, gradients, bold type, dark-first presentation, or other trends only when the approved design source, hierarchy, contrast, focus, affordance, responsive behavior, browser matrix, and performance evidence support them. Use content-visibility or contain only after checking rendering and accessibility behavior and measuring the target. Use will-change transiently and only when profiling justifies it. Choose image/font formats, dimensions, loading, fallbacks, and metrics from the delivery pipeline, browser matrix, and layout-shift evidence.
  • Tooling: run only repository-provided validation with an available dependency, target, permission, and expected evidence. Prefer documented package-manager scripts. Do not introduce implicit installs, external audits, encoding commands, or design-tool commands when prerequisites are absent.

Live Verification Rules

  • Do not add browser, API, accessibility-law, security, adoption, or version claims without checking a current authoritative source.
  • A URL is a starting point, not proof that a feature fits the repository browser matrix, product scope, build pipeline, or fallback policy.
  • Record verification date, source URL, scope, and confirmed behavior for every time-sensitive release or decision claim.
  • Use the relevant authoritative source: W3C CSS specifications for syntax, MDN for browser-feature behavior, WCAG for applicable accessibility criteria, and the official regulator for jurisdictional applicability. Check OKLCH and delivery choices against the approved pipeline and browser policy rather than claiming universal support.

Execution Order

  1. Identify repository architecture, design source, browser matrix, accessibility and legal scope, performance evidence, team conventions, tooling availability, migration cost, and security constraints.
  2. Stop for missing required evidence. Otherwise preserve token, state, semantic HTML, focus, contrast, reduced-motion, and media-geometry contracts before visual treatment.
  3. Select the smallest compatible CSS paradigm, layout, component, token, theme, media, and motion approach. Record required fallbacks, migration steps, and rejected alternatives with their concrete tradeoffs.
  4. Verify any time-sensitive version, browser, API, accessibility-law, security, adoption, or legal claim against current authoritative evidence before using it.
  5. Run available repository-native validation and report unavailable validation as evidence still needed; do not invent a successful command or result.

Output Contract

Return:

  • The recommended pattern, repository and source evidence, and why it fits the established contracts.
  • Rejected alternatives with concrete tradeoffs and migration cost where relevant.
  • Accessibility, browser, performance, security, and runtime risks, with mitigations and compatible fallbacks.
  • A fallback or migration plan, including conditions that require it.
  • Validation output: commands actually available and their results, plus missing evidence or verification still required before adoption.

Frequently asked questions

What does the Frontend Designer Lite AI skill do?

Trigger: frontend design, CSS architecture, tokens, responsive UI, accessibility, motion. Produce production UI design-system decisions without local references.

Why use Frontend Designer Lite on TypingMind?

Because you install it once and use it with any model. Frontend Designer Lite 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 Frontend Designer Lite in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/kozz36/frontend-designer-skill/tree/main/skills/frontend-designer-lite. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Frontend Designer Lite?

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 Frontend Designer Lite?

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

Is the Frontend Designer Lite AI skill free?

Yes. It is published on GitHub by kozz36 under the Apache-2.0 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 👇