Prompt Engineering Ui logo

Prompt Engineering Ui

Community
HermeticOrmus
prompt-engineering-ui

Prompt patterns for consistent UI generation. Covers precise design intent communication, component specification formats, and iterative refinement patterns for LLM-driven UI development.

Overview

PublisherHermeticOrmus
RepositoryLibreUIUX-Claude-Code
Skill nameprompt-engineering-ui
Stars
104
Forks
18
Bundled files
Instructions only
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.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

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

Installation

Install the Prompt Engineering 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.
  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/HermeticOrmus/LibreUIUX-Claude-Code.git /tmp/LibreUIUX-Claude-Code
mkdir -p .claude/skills
cp -r /tmp/LibreUIUX-Claude-Code/plugins/llm-application-dev/skills/prompt-engineering-ui .claude/skills/prompt-engineering-ui
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Prompt Engineering 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 Prompt Engineering 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 Prompt Engineering 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.

Prompt Engineering for UI Generation

Master the art of communicating design intent to LLMs. This skill covers prompt patterns specifically optimized for generating consistent, high-quality user interfaces.


When to Use This Skill

  • Writing prompts that generate consistent UI components
  • Describing design intent precisely to AI systems
  • Building reusable prompt templates for design systems
  • Iterating on UI generation with structured feedback
  • Creating few-shot examples for UI patterns
  • Debugging inconsistent UI generation outputs

The UI Prompting Challenge

UI generation is uniquely challenging because it requires:

  1. Visual precision - Exact spacing, colors, typography
  2. Behavioral specification - Interactions, states, animations
  3. Contextual coherence - Fitting within a design system
  4. Accessibility compliance - WCAG, ARIA, keyboard navigation
  5. Responsive adaptation - Multiple breakpoints, devices
  6. Code quality - Clean, maintainable output

Standard prompting techniques often fail because UI is simultaneously visual, behavioral, and technical.


Core Prompt Patterns

Pattern 1: The Component Contract

Define components as contracts with explicit input/output specifications.

markdown
## Component Contract: DataTable

### Purpose
Display tabular data with sorting, filtering, and pagination.

### Props (Inputs)
| Prop | Type | Required | Default | Description |
|------|------|----------|---------|-------------|
| data | T[] | Yes | - | Array of data objects |
| columns | ColumnDef[] | Yes | - | Column configuration |
| pageSize | number | No | 10 | Rows per page |
| sortable | boolean | No | true | Enable column sorting |
| filterable | boolean | No | false | Show filter inputs |

### Visual Specification
- **Container**: bg-white rounded-lg shadow-sm border border-gray-200
- **Header row**: bg-gray-50 text-gray-600 text-sm font-medium
- **Data rows**: hover:bg-gray-50 border-b border-gray-100
- **Typography**: Font-sans, body text 14px, headers 12px uppercase
- **Spacing**: Cell padding 12px horizontal, 8px vertical

### States
1. **Loading**: Skeleton rows with pulse animation
2. **Empty**: Centered message with icon
3. **Error**: Red border, error message below
4. **Selected**: bg-blue-50, left border accent

### Accessibility Requirements
- role="table" on container
- Sortable columns announce sort direction
- Focus visible on all interactive elements
- Keyboard navigation: Tab through headers, Enter to sort

### Output Format
React TypeScript component using Tailwind CSS.
Include JSDoc comments and prop types.

Why This Works:

  • Explicit contract eliminates ambiguity
  • Visual specs use actual CSS values
  • States prevent incomplete implementations
  • Accessibility is non-negotiable requirement

Pattern 2: Design Token Injection

Embed design tokens directly in prompts for consistency.

markdown
Generate a Card component following these design tokens:

## Tokens
```json
{
  "spacing": {
    "xs": "4px",
    "sm": "8px",
    "md": "16px",
    "lg": "24px",
    "xl": "32px"
  },
  "colors": {
    "surface": {
      "primary": "#FFFFFF",
      "secondary": "#F9FAFB",
      "elevated": "#FFFFFF"
    },
    "border": {
      "subtle": "#E5E7EB",
      "default": "#D1D5DB"
    },
    "shadow": {
      "sm": "0 1px 2px rgba(0,0,0,0.05)",
      "md": "0 4px 6px rgba(0,0,0,0.1)"
    }
  },
  "radius": {
    "sm": "4px",
    "md": "8px",
    "lg": "12px"
  }
}

Requirements

  • Card uses surface.elevated background
  • Border uses border.subtle
  • Padding uses spacing.lg
  • Border radius uses radius.lg
  • Shadow uses shadow.md

Map these tokens to Tailwind classes where possible.


**Token Mapping Strategy**:
```typescript
// Prompt can include this mapping guide
const tokenToTailwind = {
  "spacing.xs": "p-1",
  "spacing.sm": "p-2",
  "spacing.md": "p-4",
  "spacing.lg": "p-6",
  "spacing.xl": "p-8",
  "colors.surface.primary": "bg-white",
  "colors.surface.secondary": "bg-gray-50",
  "colors.border.subtle": "border-gray-200",
  "radius.lg": "rounded-xl",
  "shadow.md": "shadow-md",
};

Pattern 3: Visual Reference Chain

Chain visual descriptions from abstract to concrete.

markdown
## Component: Hero Section

### Mood (Abstract)
Confident, minimal, focused. The user should feel capable and unintimidated.

### Aesthetic (Semi-Abstract)
- Clean sans-serif typography
- Generous whitespace (40% of viewport)
- Single accent color for CTAs
- Photography: abstract, not literal

### Visual Details (Concrete)
- **Layout**: Centered, max-width 1200px, py-24
- **Headline**: text-5xl font-bold tracking-tight text-gray-900
- **Subheadline**: text-xl text-gray-600 max-w-2xl mx-auto mt-6
- **CTA Group**: mt-10 flex gap-4 justify-center
- **Primary CTA**: bg-indigo-600 hover:bg-indigo-700 text-white px-8 py-4 rounded-lg
- **Secondary CTA**: border border-gray-300 text-gray-700 px-8 py-4 rounded-lg

### Content
- Headline: "Build interfaces that inspire"
- Subheadline: "The design system that empowers creators to ship beautiful products faster."
- Primary CTA: "Get Started"
- Secondary CTA: "Learn More"

The Chain:

Mood → Aesthetic → Visual Details → Content
 ↓         ↓            ↓            ↓
Emotion   Style     CSS Values    Text

This pattern works because it builds from intention to implementation.


Pattern 4: State Machine Specification

Define component states as a state machine.

markdown
## Button Component States

### State Machine

idle → hover → pressed → idle ↓ ↓ ↓ focus focus focus ↓ ↓ ↓ disabled (terminal) loading (blocks all transitions)


### State Definitions

| State | Visual Treatment | Tailwind Classes |
|-------|------------------|------------------|
| idle | Default appearance | bg-blue-600 text-white |
| hover | Slightly darker | hover:bg-blue-700 |
| focus | Ring indicator | focus:ring-2 focus:ring-blue-500 focus:ring-offset-2 |
| pressed | Darker, slight scale | active:bg-blue-800 active:scale-[0.98] |
| disabled | Muted, no pointer | disabled:bg-gray-300 disabled:cursor-not-allowed |
| loading | Spinner, no text | Spinner SVG, opacity-50, pointer-events-none |

### Transitions
- All transitions: `transition-all duration-150 ease-in-out`
- Scale transitions: spring-like (use framer-motion if available)

### Implementation Notes
- Use `<button>` element, never `<div>`
- disabled state must be set via HTML attribute
- loading should set aria-busy="true"

Pattern 5: Constraint-First Prompting

Lead with constraints to narrow the solution space.

markdown
## Constraints (Non-Negotiable)

### Technical Constraints
- React 18+ with TypeScript strict mode
- Tailwind CSS only (no CSS-in-JS)
- No external component libraries
- Bundle size: component must be < 5KB gzipped

### Design Constraints
- Must pass WCAG 2.1 AA
- Must work without JavaScript (progressive enhancement)
- Must support RTL layouts
- Color contrast ratio >= 4.5:1

### Browser Support
- Chrome 90+, Firefox 88+, Safari 14+, Edge 90+
- No IE11 support required

### Performance Constraints
- First paint < 100ms
- No layout shift on load
- Images must be lazy-loaded

---

## Now, generate a Modal component that satisfies all constraints above.

Why Constraints First:

  • Eliminates invalid solutions immediately
  • Focuses generation on viable approaches
  • Makes review easier (checklist validation)
  • Prevents "creative" solutions that break requirements

Iterative Refinement Patterns

The Feedback Loop Protocol

Structure feedback for effective iteration:

markdown
## Iteration 1 Feedback

### What Works
- Component structure is correct
- Props interface is well-typed
- Basic styling matches tokens

### What Needs Fixing

#### Critical (Must Fix)
1. **Missing keyboard navigation**
   - Current: Only mouse interaction works
   - Required: Arrow keys to navigate, Enter to select
   - Reference: WAI-ARIA Listbox pattern

2. **Color contrast failure**
   - Current: text-gray-400 on bg-white (ratio 2.5:1)
   - Required: Minimum 4.5:1 for body text
   - Fix: Use text-gray-600 (ratio 5.7:1)

#### Important (Should Fix)
3. **Animation too fast**
   - Current: duration-75
   - Recommended: duration-150 for better perception

#### Nice to Have
4. Consider adding subtle shadow on hover

### Revised Requirements
Regenerate the component addressing Critical and Important items.

The Diff-Based Refinement

Request specific changes rather than full regeneration:

markdown
## Current Component

```tsx
<button className="bg-blue-500 text-white px-4 py-2 rounded">
  Click me
</button>

Requested Changes

  1. Add hover state: bg-blue-600 on hover
  2. Add focus ring: ring-2 ring-blue-500 ring-offset-2 on focus
  3. Add disabled state: Prop + visual treatment
  4. Add loading state: Spinner + loading prop

Output Format

Show only the modified code with inline comments explaining each change.


---

### The A/B Variant Request

Request multiple options for comparison:

```markdown
Generate 3 variants of a Card component:

## Variant A: Minimal
- No shadow
- Hairline border only
- Maximum whitespace

## Variant B: Elevated
- Pronounced shadow
- No visible border
- Subtle hover lift effect

## Variant C: Outlined
- Thick left accent border
- Light background fill
- Category color coding

## Common Requirements (All Variants)
- Same prop interface
- Same content structure
- Same responsive behavior
- Same accessibility

## Output
Provide all three variants as separate components.
Include a brief rationale for when to use each.

Few-Shot Examples for UI

Example: Button Variants

markdown
## Few-Shot Examples: Button Component

### Example 1: Primary Button
Input: Primary action button with "Submit" text
Output:
```tsx
<button className="bg-indigo-600 hover:bg-indigo-700 text-white font-medium py-2.5 px-5 rounded-lg transition-colors focus:ring-2 focus:ring-indigo-500 focus:ring-offset-2">
  Submit
</button>

Example 2: Secondary Button

Input: Secondary action button with "Cancel" text Output:

tsx
<button className="bg-white hover:bg-gray-50 text-gray-700 font-medium py-2.5 px-5 rounded-lg border border-gray-300 transition-colors focus:ring-2 focus:ring-gray-500 focus:ring-offset-2">
  Cancel
</button>

Example 3: Danger Button

Input: Destructive action button with "Delete" text Output:

tsx
<button className="bg-red-600 hover:bg-red-700 text-white font-medium py-2.5 px-5 rounded-lg transition-colors focus:ring-2 focus:ring-red-500 focus:ring-offset-2">
  Delete
</button>

Now generate: Ghost button with "Learn More" text


**Pattern Recognition**:
- Consistent class structure across examples
- Clear input-output mapping
- Similar complexity level
- Demonstrates the pattern, not just the answer

---

## Prompt Templates

### Template: Component Generation

```markdown
# Generate: {ComponentName}

## Context
Project: {ProjectDescription}
Design System: {DesignSystemName}
Framework: React + TypeScript + Tailwind

## Design Tokens
{DesignTokensJSON}

## Component Specification
Purpose: {ComponentPurpose}
Props: {PropsTable}
States: {StatesList}
Variants: {VariantsList}

## Visual Requirements
Layout: {LayoutDescription}
Typography: {TypographySpecs}
Colors: {ColorSpecs}
Spacing: {SpacingSpecs}

## Behavior
Interactions: {InteractionList}
Animations: {AnimationSpecs}
Accessibility: {A11yRequirements}

## Constraints
{ConstraintsList}

## Output
Provide production-ready React TypeScript component.
Include prop types, JSDoc comments, and usage example.

Template: Design Review

markdown
# Review: {ComponentCode}

## Review Criteria

### Design Fidelity
- Does it match the design tokens?
- Is spacing consistent?
- Are colors correct?

### Accessibility
- Keyboard navigable?
- Screen reader friendly?
- Color contrast sufficient?

### Code Quality
- Types correct?
- Props well-named?
- Logic clear?

### Performance
- Unnecessary re-renders?
- Bundle size reasonable?
- Animations performant?

## Output Format
For each criterion, provide:
- Score (1-5)
- Issues found
- Specific fixes needed

Anti-Patterns in UI Prompting

1. Vague Aesthetic Descriptions

Bad: "Make it look modern and clean" Good: "Use Inter font, 16px base, 1.5 line-height, 24px vertical rhythm"

2. Missing State Coverage

Bad: "Create a button" Good: "Create a button with idle, hover, focus, active, disabled, and loading states"

3. No Design System Context

Bad: "Use a nice blue" Good: "Use the primary color from the design tokens: #4F46E5"

4. Implicit Accessibility

Bad: "Make it accessible" Good: "Include ARIA labels, keyboard navigation per WAI-ARIA Listbox pattern, focus indicators"

5. One-Shot Expectation

Bad: Expecting perfect output on first try Good: Plan for 2-3 refinement iterations with structured feedback


Quick Reference

SituationPattern to Use
New componentComponent Contract
Ensure consistencyDesign Token Injection
Explain visual intentVisual Reference Chain
Complex interactionsState Machine Specification
Avoid reworkConstraint-First Prompting
Improving outputFeedback Loop Protocol
Minor adjustmentsDiff-Based Refinement
Exploring optionsA/B Variant Request

Integration with Other Skills

This skill pairs well with:

  • agent-orchestration/ui-agent-patterns - Prompt patterns for agent delegation
  • context-management/design-system-context - Loading tokens into prompts
  • llm-application-dev/prompt-engineering-patterns - General prompting foundations
  • design-mastery/design-principles - Visual vocabulary for descriptions

"A precise prompt is a precise thought. The UI emerges from the clarity of intention."

Frequently asked questions

What does the Prompt Engineering Ui AI skill do?

Prompt patterns for consistent UI generation. Covers precise design intent communication, component specification formats, and iterative refinement patterns for LLM-driven UI development.

Why use Prompt Engineering Ui on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/HermeticOrmus/LibreUIUX-Claude-Code/tree/main/plugins/llm-application-dev/skills/prompt-engineering-ui. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Prompt Engineering 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 Prompt Engineering Ui?

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

Is the Prompt Engineering Ui AI skill free?

Yes. It is published on GitHub by HermeticOrmus 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 👇