Building Ui

Community

MK27MK

building-ui

Design and change UI through an interactive shell. Turn a spec and a low-fidelity sketch, or an existing component, into a high-fidelity mock.html in the project's own style, iterate on it with the user, then implement it. Or open the shell over the running app, where the user inspects, comments on and restyles real elements, and the changes go straight into code and tests. Use when the user invokes /building-ui, asks to mock up a new component or UI element from a sketch, asks to change a component through a mock, asks to implement an approved mock, or asks to review or change the UI of the running app.

Overview

Publisher	MK27MK
Repository	`uxed`
Skill name	`building-ui`
Stars	80
Forks	3
Bundled files	13
Links	Skill source Repository

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.
13 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by MK27MK on GitHub. Read the source before you install it.

Installation

Install the Building Ui AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

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.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/MK27MK/uxed/tree/main/skills/building-ui
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.

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/MK27MK/uxed.git /tmp/uxed
mkdir -p .claude/skills
cp -r /tmp/uxed/skills/building-ui .claude/skills/building-ui

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Building Ui 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 Building Ui 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 Building Ui 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.

Building UI

The shell works in two modes: App and Mock. Read only the file for the phase you are in, and the sections below that it points to.

Mock

To work on a component in isolation. Use it for a new component, or when the user explicitly asks to change a component:

Phase	When	File
Mock	No mock exists for the component yet.	mock.md
Review	A mock exists and the user has not approved it.	review.md
Implement	The user approved the mock and asks to implement it.	implement.md

The flow pauses when the user approves the mock. Implement the component only when the user asks.

App

For changes to UI elements the running app already shows: app.md.

Pick mock mode for a new component or when the user explicitly asks to change a component, and app mode for any other change to existing UI. If a request fits both, ask.

Conventions

notes.mjs is node scripts/notes.mjs. It reads and updates the notes for both modes: run it with no arguments for its commands. Prefer it to reading or editing the notes JSON yourself. <notes> is the notes file of the mode you are in.

Paths starting with assets/ or scripts/ are relative to this skill's folder.

Open the shell

Once the mode's file has started scripts/serve.mjs in the background, which prints its URL on port 4870 or the next free one:

Open the page the mode's file names in the browser. Prefer the built-in browser pane when it is available.
Check once that the shell's toolbar shows and the console has no errors from /__mock/ routes or, in mock mode, from mock.html. Do no further visual QA.
End the turn. Do not explain the shell in chat: its ? button, at the bottom left, shows how to use it.

Each time the user hands back the turn

Run notes.mjs pending <notes>. It lists the edits, renames and comments left to handle.
Edits. Write each style edit into the element's CSS, and each text edit into the matching instance. Then run notes.mjs clear <notes> edits <id>….
Renames. Apply each rename where the mode's file says.
Comments. Address each listed comment. A change in look (style, layout, spacing, wording) is an edit: make it now, with no rule. A change in behaviour is a rule: add it with notes.mjs rule <notes> <id> "<text>", or, only when it has several readings that lead to different changes, add the one you would pick with --assumption, so the user can confirm it. When you cannot tell what a comment targets, ask in the reply and change nothing. Then reply with notes.mjs reply <notes> <id> "<text>": say what changed, or ask a question. Never resolve comments: only the user resolves them.
Rules. Treat every rule the user wrote, changed or confirmed as spec, and make the UI obey it. Only the user turns an assumption into spec: never change a rule's source yourself, unless the user settles it in a comment or in chat.
Apply any changes the user asked for in chat.
The page reloads itself when the notes or the UI change. If the user has unsaved edits, the page shows a toast instead; tell them to reload before they edit again, because a stale page overwrites your notes on its next save.
Report in chat only what the comment replies do not already say: changes the user asked for in chat, questions, and blockers. Do not restate your replies.

Repeat until the user is done, then stop the server and finish as the mode's file says.

Rename in code

An element's name in code is its component, its CSS-module class as stem.local, a plain class, or its data-ui value. Rename the class in its stylesheet and every use, or the data-ui value. A new stem means renaming the stylesheet file and its imports, and the component too when they share the name. A renamed component also renames its folder to the new name in kebab-case, and its files the way the project names them.

Implement rules test-first

Implement an element only when all its rules are spec. Leave elements with assumptions for the user to confirm in the shell: tell the user that assumptions are left to confirm there, and do not list or ask about them in chat.
Run the test-first loop with the project's TDD skill. If the project has none, stop before writing code and ask the user with AskUserQuestion, offering three options:
- Stop for now (Recommended): end the turn so the user can add a TDD skill.
- Continue with TDD, without a skill: write each test and watch it fail before you implement it.
- Continue without TDD: implement first, then write the tests.
Each new or changed rule becomes at least one test, and each removed rule a test that the behaviour is gone.
In code, use the project's token wherever a token has exactly the value you would write.

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 Building Ui AI skill do?

Why use Building Ui on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/MK27MK/uxed/tree/main/skills/building-ui. 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 Building Ui?

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 Building Ui?

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

Is the Building Ui AI skill free?

It is published on GitHub by MK27MK. 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

Solid

openclaw

OrganizationPopular

Write, port, review, or test Solid 2 UI in OpenClaw (Control UI, plugin views, Canvas hosts), including state ownership, Lit interop during the migration, native overlays, testing, performance, and…

Canvas

Present hosted widget documents on a connected macOS panel and control panel visibility or navigation.

Diagram Maker

Create SVG/HTML or Excalidraw diagrams for concepts, architecture, flows, and whiteboards.

Grill With Docs

A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Building UI

Mock

App

Conventions

Open the shell

Each time the user hands back the turn

Rename in code

Implement rules test-first

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now