Ui Web logo

Ui Web

Community
rshzrh
ui-web

Generate slop-free UI for React + Tailwind + shadcn/ui. Use whenever the user asks to build, design, scaffold, restyle, or improve a screen, component, page, landing, app, or any web interface. Replaces frontend-design, design-html, design-shotgun, design-consultation. Two phases — tokens first (with approval gate), then UI — both gated by a self-audit checklist that the model MUST run before claiming done.

Overview

Publisherrshzrh
Repositoryui-design-skills
Skill nameui-web
Stars
5
Forks
0
Bundled files
10
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.

  • 10 bundled files

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

  • Open source

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

Installation

Install the Ui Web 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/rshzrh/ui-design-skills/tree/main/skills/ui-web
  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/rshzrh/ui-design-skills.git /tmp/ui-design-skills
mkdir -p .claude/skills
cp -r /tmp/ui-design-skills/skills/ui-web .claude/skills/ui-web
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ui Web 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 Ui Web 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 Ui Web 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.

ui-web — Slop-Free Web UI Skill

You are generating production-grade web UI in React + TypeScript + Tailwind. This skill exists because untouched LLMs produce AI slop: purple→blue gradients, Inter everywhere, 3-icon-card grids, "Lightning fast / Powered by AI" copy, glassmorphism by default, arbitrary spacing values, and the median of every Tailwind tutorial scraped from GitHub. This skill replaces guesses with constraints.

The non-negotiable rule: NEVER use a buzzword in place of a token. Words like clean, modern, professional, sleek, premium are banned from your reasoning. Every aesthetic choice must trace back to a named anchor recipe in anchors.md and a token in the project's tailwind.config.ts.


Workflow (follow in order)

Step 0 — Project scan (decides bootstrap vs extend mode)

Before any other action, check for:

DESIGN_RULES.md            # at repo root → if exists, you are in EXTEND mode
tailwind.config.{ts,js}    # → if exists with custom theme, you are in EXTEND mode
app/globals.css            # check for CSS custom properties / tokens
src/styles/tokens.css
  • If DESIGN_RULES.md exists: read it. Read the existing tailwind.config.*. Skip to Step 3 (Phase 2: UI). Use only existing tokens. NEVER add new colors/spacing/radii without asking.
  • If tokens exist but no DESIGN_RULES.md: read tokens, infer the closest matching anchor, ask the user to confirm the inferred anchor, then write DESIGN_RULES.md and proceed to Phase 2.
  • If neither exists: you are in bootstrap mode. Continue to Step 1.

Step 1 — Intake (interactive, bootstrap only)

Use AskUserQuestion to gather, in this order:

  1. Anchor pick — show the 8 recipes from anchors.md (one-line summary each). Let the user pick 1–3. If 2–3, ask which is primary and which contributes type vs color vs motion.
  2. Ban menu — show the 8 ban patterns from ban-patterns.md with the 3 default-on items pre-checked. Confirm.
  3. Project context:
    • Content density: airy / balanced / dense
    • Primary surface: marketing site / app / docs / dashboard / mobile-web
    • Dark mode: required / optional / none
    • Brand color (if any): hex value or "derive from anchor"

Do NOT ask vibe questions ("how should it feel?"). Ask only the questions above.

Step 2 — Phase 1: tokens (gated)

Generate, in this order:

  1. tailwind.config.ts — extend theme using the chosen anchor's recipe. Use the template at templates/tailwind.config.ts.tmpl. Required keys: colors, spacing, borderRadius, fontSize, fontFamily, boxShadow, transitionDuration, transitionTimingFunction.
  2. app/globals.css (or src/index.css for Vite) — CSS custom properties mirroring the Tailwind tokens, plus dark-mode overrides if requested.
  3. eslint.config.js patch — enable the slop-blocking lint rules from templates/eslint.slop.cjs.tmpl. Add a lint:design script to package.json.
  4. DESIGN_RULES.md at repo root — fill templates/DESIGN_RULES.md.tmpl with the chosen anchors, active bans, token file paths, and verification command.
  5. CLAUDE.md patch — append (do not overwrite): Before any UI/UX work, read and follow DESIGN_RULES.md.

Then run the Token Audit from verification.md and print the PASS/FAIL table to the user. Do not proceed to Phase 2 until the user approves.

Step 3 — Phase 2: UI

Generate the requested screen/component using only tokens from the config. Follow:

  • component-anatomy.md — exact rules for button, input, card, list, nav, modal, toast, empty-state, form-field, table-row.
  • motion.md — easing curves and durations.
  • copy-voice.md — every string is real, contextual copy. No lorem ipsum, no filler.
  • icons-imagery.md — icon library, sizing, alignment. No emoji bullets in serious UI.

Then run the UI Audit from verification.md AND pnpm lint:design (or npm/yarn equivalent). Print the audit table. Fix every FAIL before claiming done.


Files in this skill

FileWhen to read
anchors.mdStep 1, before showing the user choices. ALWAYS read fully — never quote from memory.
ban-patterns.mdStep 1, when showing ban menu. Step 2 and 3, when verifying.
component-anatomy.mdStep 3, before writing any component.
motion.mdStep 3, when adding any transition or animation.
copy-voice.mdStep 3, before writing any user-facing string.
icons-imagery.mdStep 3, when choosing icons or images.
verification.mdAfter Step 2 (Token Audit) and after Step 3 (UI Audit). MANDATORY — do not skip.
templates/*Use as the basis for the generated files; fill placeholders, never ship as-is.

Hard rules (the model must obey these always)

  1. No arbitrary values. Forbidden: mt-[13px], w-[372px], text-[#3a4f6b], rounded-[7px]. Every value must be a token in tailwind.config.ts. If you need a value that does not exist, ask the user to add it to the token set.
  2. No buzzwords in code or comments. Banned strings in your output: clean, modern, professional, sleek, premium, beautiful, elegant (except in copy if the user explicitly asked for those words).
  3. No default-style-tells unless the chosen anchor explicitly allows them. Specifically: no bg-gradient-to-r from-purple-* to-pink-*, no from-indigo-* to-purple-*, no backdrop-blur on regular cards (only on iOS-style chrome/sheets if anchor allows).
  4. No filler copy. No lorem ipsum. No "Lightning fast", "Built for the modern web", "Powered by AI", "Get started in seconds", "The future of X", "Beautiful and intuitive".
  5. Touch targets ≥ 44px on any control intended for tap.
  6. Focus rings always visible. Never focus:outline-none without an explicit replacement.
  7. Every async surface has empty + loading + error states. No exceptions.
  8. Dark mode parity. If the project supports dark mode, every component you write must work in both — verify by reading the colors you used.
  9. You must run the verification checklist before responding "done". Print the audit table to the user.

What this skill is NOT

  • Not a magic-aesthetic generator. You still have to pick an anchor and follow it.
  • Not a copy generator. Phase 2 stops and asks for real copy when needed.
  • Not a brand-design tool. Brand identity (logo, custom typography family selection beyond what anchors provide) is the user's call.
  • Not a replacement for design review with humans. The audit table is a floor, not a ceiling.

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 Ui Web AI skill do?

Generate slop-free UI for React + Tailwind + shadcn/ui. Use whenever the user asks to build, design, scaffold, restyle, or improve a screen, component, page, landing, app, or any web interface. Replaces frontend-design, design-html, design-shotgun, design-consultation. Two phases — tokens first (with approval gate), then UI — both gated by a self-audit checklist that the model MUST run before claiming done.

Why use Ui Web on TypingMind?

Because you install it once and use it with any model. Ui Web 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 Ui Web in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/rshzrh/ui-design-skills/tree/main/skills/ui-web. 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 Ui Web?

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 Ui Web?

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

Is the Ui Web AI skill free?

Yes. It is published on GitHub by rshzrh 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 👇