Ui Design logo

Ui Design

OrganizationPopular
EpicenterHQ
ui-design

Design and review Epicenter interfaces while collapsing local UI primitives into the shared component system. Use when co-designing, exploring comparable apps, turning a feature idea into Svelte UI, or changing `packages/ui`, not for a tiny CSS repair.

Overview

PublisherEpicenterHQ
Repositoryepicenter
Skill nameui-design
Stars
4.8K
Forks
386
Bundled files
5
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.

  • 5 bundled files

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

  • Open source

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

Installation

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

Use it in TypingMind

Enable Ui Design 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 Design 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 Design 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 Design

ui-design preserves the user-visible objective while reimagining the simplest greenfield composition and collapsing local markup, arbitrary styling, and duplicated visual contracts into @epicenter/ui, except where app-specific behavior or a load-bearing interaction earns the custom shape.

The existing screen is evidence, not a structure to reproduce. For Epicenter product apps, @epicenter/ui is the default system. Treat shadcn-svelte as implementation lineage and upstream grounding, not as an app import path.

Canonical Path

  1. Name the user-visible objective in one concrete sentence. When the product direction is unsettled, read references/direction-discovery.md before choosing the composition.
  2. Read the current and neighboring surfaces. Inventory the workflow, hierarchy, accessibility, important states, and recognizable product intent that matter.
  3. Sketch the simplest greenfield composition using the natural anatomy of existing @epicenter/ui primitives.
  4. Run the component-system collapse pass below. Refuse exact reproduction of incidental markup, spacing, colors, breakpoints, and one-off states.
  5. Implement the whole state surface: loading, empty, error, disabled, pending, selected, and filter-empty where relevant.
  6. Verify visible, responsive, and interactive results in the browser.
  7. Run UI pre-flight before final handoff or committing the change.

For substantial work, state the design read:

txt
Objective: <what the user accomplishes>.
Greenfield shape: <the simplest natural component-system composition>.
Collapse: <the local structure or styling that disappears>.

Component-System Collapse Pass

Run this for every meaningful UI implementation or review, not only work called a redesign. This is the UI application of asymmetric-wins; load that skill when a candidate refusal needs an explicit product-loss versus deletion-prize decision.

txt
Preserve the product sentence.
Challenge the inherited composition.
Refuse the incidental visual promise.
Delete the local code family.

Inspect every wrapper, class bundle, arbitrary Tailwind value, custom color, locally reconstructed state surface, and primitive override. Ask:

  1. Does an existing primitive or variant already own this contract?
  2. Can the primitive's natural shape replace the wrapper and its classes?
  3. Is the exact local detail load-bearing for comprehension, accessibility, interaction, brand, or required geometry?
  4. If it survives, who owns it: the app's product behavior or the shared system?

Implement obvious collapses directly. Ask the user only when the refusal changes workflow, hierarchy, product meaning, brand posture, or a plausibly load-bearing detail.

Keep custom UI app-local only when it owns app-specific behavior or product meaning. Move a stable visual contract into packages/ui when several apps need it or it owns an accessibility or interaction contract.

Human Taste Gate

Ask one focused question when human preference would materially change the visual language, brand posture, primary hierarchy, navigation, density, or the direction of an expressive public surface. Present concrete alternatives and a recommendation. Do not ask about choices settled by the shared system, accessibility, existing product intent, or a mechanical collapse.

Product UI And Expressive Surfaces

Product UI should usually be dense, quiet, and operational. Distinctiveness comes from workflow fit, hierarchy, copy, state design, and purposeful details. Marketing pages, docs entry points, prototypes, and explicitly expressive work can use a stronger visual identity, but still run the collapse pass and keep only details that serve the chosen direction.

References

  • Read references/direction-discovery.md when the user wants to co-design, compare directions, or turn a rough UI idea into a buildable surface.
  • Read references/comparable-apps.md when the design question depends on category fit or what other apps do.
  • Read references/component-system.md for every meaningful Epicenter product UI implementation or review. It owns collapse smells and exceptions, then routes exact component and package mechanics to the authoritative packages/ui guide.
  • Read references/anti-slop-tells.md for public, marketing, docs, portfolio-like, or highly polished work where templated output is a material risk.
  • Read references/preflight.md before final handoff or committing any visible UI change. Report only failures or intentional exceptions.

Delegation Boundaries

  • styling owns CSS, Tailwind, spacing, wrappers, overflow, and scroll traps.
  • svelte owns component structure, runes, lifecycle, and state mechanics.
  • tanstack-table owns table state, columns, sorting, and row identity.
  • writing-voice owns visible interface copy and tone.

For an explicit external standards or accessibility audit, WebFetch the Vercel Web Interface Guidelines and apply them directly, reporting findings as file:line. Fetch them per review rather than caching a copy here, because the upstream rules change and a stale copy would be worse than no copy.

txt
https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md

The design decision and component-system collapse stay here even when another skill owns their implementation mechanics.

Final Output

Report the changed surfaces, browser verification, any material deletion prize, and only the custom shapes or pre-flight exceptions that intentionally remain.

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 Design AI skill do?

Design and review Epicenter interfaces while collapsing local UI primitives into the shared component system. Use when co-designing, exploring comparable apps, turning a feature idea into Svelte UI, or changing `packages/ui`, not for a tiny CSS repair.

Why use Ui Design on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/EpicenterHQ/epicenter/tree/main/.agents/skills/ui-design. 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 Design?

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 Design?

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

Is the Ui Design AI skill free?

It is published on GitHub by EpicenterHQ. Check the repository for licensing terms. 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 👇