Open Claude Ui Design logo

Open Claude Ui Design

Community
maxritter
open-claude-ui-design

Create or redesign product UI in an existing codebase, including a wireframe, substantive visual variations, and an interactive prototype. Use for requests about UI layout, visual direction, screens, flows, components, a landing page, dashboard, or making an interface feel polished. Do not use for backend work, system architecture, decks, or logic-only UI fixes.

Overview

Publishermaxritter
Repositoryopen-claude-design
Skill nameopen-claude-ui-design
Stars
38
Forks
4
Bundled files
7
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.

  • 7 bundled files

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

  • Open source

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

Installation

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

Use it in TypingMind

Enable Open Claude 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 Open Claude 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 Open Claude 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

Create intentional product interfaces inside the user's real codebase. The current product is the source of truth; design expertise improves it without replacing its language by reflex.

Important boundaries

  • Inspect the existing design system, components, tokens, theme, screenshots, assets, and neighboring screens before proposing visual changes.
  • A non-visual logic change in a UI file is not a design task. Preserve the rendered result and handle it directly without expanding scope.
  • Work in the project's native framework and file structure. Do not replace an application with a standalone HTML mockup unless the user explicitly requests an isolated prototype.
  • Reuse real content and assets. Mark a missing asset with a labeled placeholder; do not invent product claims, statistics, testimonials, features, or destinations.
  • When Impeccable is also available, this skill owns the overall product-design workflow. Use Impeccable only for a requested named refinement, supporting agent, hook evidence, or detector result; do not run a second competing design process.

Direction count

Create one strong, complete direction by default. A direct request to design, redesign, or implement a surface does not imply comparative exploration. Use multiple variants only when the user explicitly asks for options, alternatives, or comparison, or when a materially unresolved product direction makes side-by-side evidence cheaper than a clarifying question. An established product system plus a clear objective normally has one product-consistent answer.

Scope

The user's explicit surface is the boundary: do not narrow a requested full-screen redesign or expand a requested region. If the request names a whole surface but gives only a localized change objective and the intended breadth is materially ambiguous, ask one scope question. When no answer is available, design the named change and leave the surrounding UI unchanged and omitted.

Before sketching or specifying, separate provided product facts, verified existing UI, and unknowns. Missing evidence removes authority; it never permits “generic” or “illustrative” fields, settings, sections, actions, policies, or product consequences, and an unknown never becomes plausible-looking interface copy.

For a partial-surface request, start the deliverable with Scope: <named region> only; surrounding UI unchanged and omitted. Show only that region: a full-page wireframe, a named surrounding section, or an invented neighboring action is out of scope. Render an unknown high-stakes consequence as a bracketed placeholder such as [verified deletion consequence], never as factual copy. A missing repository never expands scope; in a self-contained brief, the brief is authoritative.

Select the relevant procedure

Read only the references needed for the request, resolved relative to this skill:

RequestRequired reference
New UI, redesign, or ambiguous visual directionreferences/discovery-and-direction.md
Greenfield work with no brand, system, or references to inheritreferences/directions.md
A common surface type: marketing section, application shell, table, form, settings, dashboard, auth page, or commerce screenreferences/layout-patterns.md
Explicit wireframes, alternatives, or materially unresolved visual directionreferences/exploration.md
Clickable prototype, interaction demo, or new flowreferences/prototype.md

Combine references only when the task spans their modes. A redesign that asks for three interactive alternatives uses discovery, exploration, and prototype; a focused styling change may need none beyond this router.

Execution contract

  1. Ground every decision in inspected product evidence or an explicit greenfield direction.
  2. Ask only when the answer materially changes audience, scope, brand, flow, fidelity, or the axes of comparison. Make reversible local decisions autonomously and state the consequential ones.
  3. Build the smallest complete surface that proves the direction, then extend it consistently through components and tokens. On a multi-section page, finish and check one section against the product system and the craft rules before starting the next; one review pass at the end hides the defects the early sections introduced.
  4. Verify the actual interface with the repository's named browser, simulator, or UI driver: interact with the primary path, re-snapshot the result, inspect representative widths and supported themes, and report unverified states.
  5. For a final visual audit or polish request, use the open-claude-ui-review skill so review findings, accessibility, and fixes have one owner.

When the user requests an outline or plan and explicitly forbids implementation, include a Verification plan in the delivered artifact. It names the primary interaction, one representative failure, keyboard operation, the viewport/theme matrix, and the state transitions that must be re-snapshotted after implementation.

Every asynchronous action in a designed flow includes a retryable failure state: keep relevant input/context, show an actionable error, re-enable the action when safe, and define where focus moves.

Completion

The result is complete when the requested surface works in the real application, follows either the existing system or a documented new direction, contains no invented scope, and has interaction plus visual evidence at the affected states and viewports.

When the task uses an existing canonical Claude Design product model, finish by integrating the accepted refinement into that model through open-claude-design's model-consolidation workflow. Check the integrated screen and its neighboring navigation, then retire authorized temporary previews after verified merge-back. Keep one current model per product instead of leaving the new screen as a competing standalone deliverable. Honor a user's request to keep an exploration separate and preserve unrelated pages and edits.

When not to use

  • Backend, API, database, infrastructure, or software-architecture design
  • Slide decks, presentations, print documents, or generic image creation
  • Logic-only fixes in TSX/JSX or other UI files whose rendered behavior must remain unchanged
  • Design-system extraction without a requested screen or flow; use open-claude-design-system
  • Report-only UI audits or pre-ship polish; use open-claude-ui-review
  • Accessing, importing, or changing a Claude Design project; use open-claude-design, then return here for local adaptation

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 Open Claude Ui Design AI skill do?

Create or redesign product UI in an existing codebase, including a wireframe, substantive visual variations, and an interactive prototype. Use for requests about UI layout, visual direction, screens, flows, components, a landing page, dashboard, or making an interface feel polished. Do not use for backend work, system architecture, decks, or logic-only UI fixes.

Why use Open Claude Ui Design on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/maxritter/open-claude-design/tree/main/skills/open-claude-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 Open Claude 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 Open Claude Ui Design?

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

Is the Open Claude Ui Design AI skill free?

It is published on GitHub by maxritter. 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 👇