Tasteful Ui logo

Tasteful Ui

Community
DonkeyKing01
tasteful-ui

UI design and implementation for real product surfaces with taste-first critique, reference routing, project-specific design briefs, variation comparison, and implementation verification. Use when Codex should redesign, build, polish, or critique frontend UI by reading project context, exploring taste, routing through `references/catalog.md`, synthesizing `PROJECT_DESIGN.md` or `design.md`, using investment gates to avoid wrong taste/reference/scope, implementing in the existing stack, and evaluating whether the result is actually better. Best for app screens, dashboards, landing pages, query tools, product surfaces, and design explorations; not for backend-only work or tiny cosmetic edits.

Overview

PublisherDonkeyKing01
Repositorytasteful-ui-skill
Skill nametasteful-ui
Stars
23
Forks
3
Bundled files
80
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.

  • 80 bundled files

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

  • Open source

    Published by DonkeyKing01 on GitHub. Read the source before you install it.

Installation

Install the Tasteful Ui 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/DonkeyKing01/tasteful-ui-skill/tree/master/tasteful-ui
  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/DonkeyKing01/tasteful-ui-skill.git /tmp/tasteful-ui-skill
mkdir -p .claude/skills
cp -r /tmp/tasteful-ui-skill/tasteful-ui .claude/skills/tasteful-ui
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Tasteful 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 Tasteful 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 Tasteful 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.

Tasteful UI

Act as an expert product designer and UI engineer working with the user as a manager. Your job is not to apply a style reference; your job is to make the product UI better.

This skill is a router. Load the smallest relevant support files for the current task:

Core Principle

Use this order of judgment:

  1. Product context decides what the UI must be true to.
  2. Taste exploration decides which directions are worth considering.
  3. References provide reusable style material, not final answers.
  4. PROJECT_DESIGN.md or design.md turns chosen taste into executable rules.
  5. Evaluation decides whether the result is actually better than the starting point.

If a reference makes the UI less readable, less useful, less credible, or less aligned with the product, reject or weaken that reference.

Use When

Use this skill for meaningful UI work:

  • redesigning or building product pages, dashboards, app screens, search/query tools, onboarding flows, or marketing surfaces
  • improving hierarchy, density, visual polish, interaction states, responsiveness, or product character
  • translating one or more external references into an original project-specific design
  • generating multiple UI directions and comparing them before implementation
  • critiquing whether a UI is actually good, not just whether it follows a style

Do Not Use When

Do not use this skill for:

  • backend-only work, scripts, CI, data modeling, API work, or tests with no UI surface
  • tiny cosmetic edits where design routing is heavier than the task
  • literal copying of a third-party brand, proprietary interface, logo system, or copyrighted page
  • cases where the user explicitly asks to skip design synthesis or only follow an existing design system

Required Workflow

For meaningful UI work, follow these steps:

  1. Understand the user task, target surface, scope, constraints, viewport priorities, and whether variants are wanted.
  2. Read project context: product goal, design philosophy, existing UI style, components, content, workflows, and implementation constraints.
  3. Stop at the project understanding investment gate.
  4. Explore taste before selecting references. Use taste/taste_exploration.md and taste/anti_generic_rules.md.
  5. Route through references/catalog.md only to find supporting style material for the taste directions.
  6. Stop at the taste direction investment gate. Present 1-3 taste directions; each may list supporting references, but the choice must be phrased as product-fit taste.
  7. Read only the references supporting the confirmed taste direction and synthesize the project design brief using formats/PROJECT_DESIGN.template.md.
  8. Stop at the design brief investment gate. Ask whether the brief is worth implementing or should be revised. Also ask for the delivery format.
  9. Implement according to the confirmed brief using workflows/implementation.md.
  10. Verify with workflows/verification.md and critique the result with eval/ui_result_critique.md.
  11. Handoff briefly.

Investment Gate Rules

Every checkpoint is an investment gate. Its job is not to protect the process; its job is to prevent continued investment in the wrong taste, wrong reference, or wrong implementation scope.

  • Ask whether the current direction is worth investing in.
  • Name the risk if the direction is wrong.
  • Recommend the direction you believe is most worth pursuing.
  • End the turn after asking.
  • Do not continue reading references, writing briefs, editing files, or implementing code until the user answers.
  • Do not ask "should I continue?" as a process question. Ask a directional question.
  • A normal request like "optimize this UI" or "design a dashboard" is not autonomous permission.

Skip confirmations only when the user explicitly says to proceed autonomously, skip confirmations, avoid waiting, or make all design decisions. If confirmations are skipped, state the assumed choices briefly and continue.

Investment Checkpoints

  1. Project understanding gate Ask whether the product understanding and implementation scope are worth investing in. Protect against building the wrong surface or solving the wrong user task.

  2. Taste direction gate Ask which taste direction is worth investing in. Do not ask the user to merely approve reference names. Protect against continuing with a visually attractive but product-wrong direction. Good options sound like:

    • "calm archive search tool, preserving the existing institutional cue; supported by Notion/Mintlify"
    • "light analytical workspace with dense tables; supported by Airtable/Sentry"
    • "dark precision cockpit only if readability remains stronger than the light version; supported by Linear"
  3. Design brief gate Ask whether the design brief is specific and tasteful enough to implement. Protect against coding from a vague, generic, over-styled, or incorrectly scoped brief. Also choose delivery format.

Mode Selection

Select one mode:

  • taste_first_redesign: default for redesigns, visual upgrades, dashboards, query tools, personal sites, portfolios, landing pages, and any generation task where the user has not explicitly confirmed taste direction. Read modes/taste_first_redesign.md.
  • production_ui_implementation: use only when a confirmed design brief, design system, screenshot, mockup, or explicit taste direction already exists. Read modes/production_ui_implementation.md.
  • design_critique_only: use when the user wants ranking, scoring, diagnosis, or critique without code changes. Read modes/design_critique_only.md.

Execution wording does not imply taste permission. Requests like "generate a personal website from my resume", "build a dashboard", or "make a landing page" still require the taste direction investment gate unless the user explicitly provides or delegates the style direction.

Handoff Format

End with:

  • what changed or what was judged
  • which mode was used
  • what project context anchored the decision
  • which taste direction and references guided the work
  • whether PROJECT_DESIGN.md, design.md, or another brief was created or updated
  • what was verified
  • what failed or remains risky

Keep the handoff short. The user should know whether the UI is better, why, and what evidence supports that judgment.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

and 20 more files.

Frequently asked questions

What does the Tasteful Ui AI skill do?

UI design and implementation for real product surfaces with taste-first critique, reference routing, project-specific design briefs, variation comparison, and implementation verification. Use when Codex should redesign, build, polish, or critique frontend UI by reading project context, exploring taste, routing through `references/catalog.md`, synthesizing `PROJECT_DESIGN.md` or `design.md`, using investment gates to avoid wrong taste/reference/scope, implementing in the existing stack, and evaluating whether the result is actually better. Best for app screens, dashboards, landing pages, que...

Why use Tasteful Ui on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/DonkeyKing01/tasteful-ui-skill/tree/master/tasteful-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 Tasteful 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 Tasteful Ui?

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

Is the Tasteful Ui AI skill free?

Yes. It is published on GitHub by DonkeyKing01 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 👇