agentleFS
ryancproductryancproduct/product-ai-toolkit/skills/rapid-prototype/SKILL.md

rapid-prototype

Skill4 starsChanged 6 months agoMITView on GitHub

Build UI prototypes using the zero-build HTML system $ARGUMENTS

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Update <title> and main heading

    line 57
  2. Distinctive:

    Add any required JavaScript interactions

    line 59
  3. Distinctive:

    Add Interaction Engineer's <script> block

    line 164
  4. Distinctive:

    Empty States: - No data views

    line 203
  5. Distinctive:

    Before delivering prototype, verify:

    line 329
  6. Distinctive:

    If user provides a Figma URL, use Figma MCP tools:

    line 344

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Rapid Prototype Skill

Overview

Build interactive UI prototypes in seconds using a zero-build HTML prototyping system. No npm, no webpack, no waiting. Edit → Save → Refresh = 2 seconds.

Source Kit Location: [YOUR_PROTOTYPE_TEMPLATES_PATH]/

When to Use This Skill

  • Creating interactive mockups for stakeholder review
  • Validating UI concepts quickly
  • Testing user flows before implementation
  • Building throwaway demos for presentations
  • Documenting design patterns

Invocation

/rapid-prototype "Create a user management screen with search, add button, and user list with actions"

Or natural language:

  • "Build a prototype for..."
  • "Create a quick mockup of..."
  • "Prototype the..."

Workflow: Single Agent (Default)

For straightforward prototypes (1-3 sections, <10 components):

Step 1: Parse Requirements

Extract from the user's brief:

  • Page title - What is this screen called?
  • Components needed - Buttons, cards, lists, forms, tabs, modals?
  • Layout structure - Header? Sidebar? Main content sections?
  • Interactivity - Tabs? Modals? Filters? Form validation?

Step 2: Read Source Materials

Read: [YOUR_PROTOTYPE_TEMPLATES_PATH]/blank_prototype_template.html
Read: [YOUR_PROTOTYPE_TEMPLATES_PATH]/DesignSystem.html

Step 3: Assemble Prototype

  1. Copy blank_prototype_template.html structure
  2. Update <title> and main heading
  3. Replace main content area with assembled components
  4. Add any required JavaScript interactions
  5. Save to user's working directory

Step 4: Deliver

# macOS - Open in browser
open [prototype_filename].html

Provide user with:

  • File path
  • Key interactions to test
  • Suggested iterations

Workflow: Parallel Agents (Complex Prototypes)

For complex prototypes (4+ sections, 10+ components, multiple interactions):

Spawn parallel agents using Task tool with these specialized roles:

Agent 1: Layout Architect

Prompt:

You are building a prototype. Your job: CREATE THE PAGE STRUCTURE.

Read: [YOUR_PROTOTYPE_TEMPLATES_PATH]/blank_prototype_template.html

Task: Build the HTML skeleton for [DESCRIBE LAYOUT]
- Keep the full <head> section with Tailwind config
- Keep header and sidebar from template
- Create empty placeholder sections in <main> with comments like:
  <!-- SECTION: User List -->
  <!-- SECTION: Filter Bar -->
- Do NOT add component details yet

Output: The complete HTML file with placeholder sections.

Agent 2: Component Extractor

Prompt:

You are extracting components from the design system.

Read: [YOUR_PROTOTYPE_TEMPLATES_PATH]/DesignSystem.html

Task: Extract and adapt these components: [LIST COMPONENTS]

For each component:
1. Find it in DesignSystem.html
2. Copy the markup
3. Adapt content/labels for the use case
4. Ensure focus-ring class on all interactive elements

Output: A markdown document with each component's adapted HTML, ready to paste.

Agent 3: Interaction Engineer

Prompt:

You are adding JavaScript interactivity to a prototype.

Reference patterns from: [YOUR_PROTOTYPE_TEMPLATES_PATH]/CLAUDE.md

Task: Write vanilla JavaScript for: [LIST INTERACTIONS]
- Tabs switching
- Modal open/close
- Form validation
- Filter/search functionality
- Checkbox selection states

Rules:
- Vanilla JavaScript only (no frameworks)
- Use event delegation for lists
- All JS goes at end of <body> in single <script> block

Output: Complete <script> block ready to paste.

Agent 4: Accessibility Validator

Prompt:

You are validating accessibility for a prototype.

Reference: [YOUR_PROTOTYPE_TEMPLATES_PATH]/CLAUDE.md (Accessibility section)

Task: Review this HTML and provide fixes:
[PASTE ASSEMBLED HTML]

Check for:
- focus-ring class on ALL interactive elements
- ARIA attributes (aria-label, aria-current, aria-expanded)
- Semantic HTML (<nav>, <main>, <header>, proper heading hierarchy)
- Alt text on images (empty alt="" with aria-hidden="true" for decorative)
- Skip link at top of page

Output: List of issues with corrected HTML snippets.

Assembly (After Parallel Agents Complete)

  1. Take Layout Architect's skeleton
  2. Insert Component Extractor's components into placeholder sections
  3. Add Interaction Engineer's <script> block
  4. Apply Accessibility Validator's fixes
  5. Save and test

Component Reference (Quick Lookup)

Navigation

  • Breadcrumb - Page hierarchy trail
  • Link - Text links with optional icons
  • Header - Top navigation bar
  • Tabs - Bordered or pill variants
  • Menu - Action menus (Edit, Delete, etc.)

Forms

  • Input - Text, search, number variants
  • Checkbox - Single or group
  • Radio - Radio button groups
  • Form Field - Label + input + helper text
  • Date Picker - Date/time selection
  • File Upload - File input with preview

Actions

  • Button - Primary, Secondary, Tertiary, Destructive
  • Action Bar - Bulk action toolbar
  • Filter Button - Filter toggle
  • Filter Tag - Active filter chip

Display

  • Card - Basic, interactive, stat card
  • List Item - Entity rows with avatars
  • Avatar - User/entity images
  • Badge - Status indicators
  • Type Label - Category labels

Feedback

  • Banner - Info/warning/error messages
  • Dialog - Modal dialogs
  • Empty States - No data views

Layout

  • Accordion - Collapsible sections
  • Carousel - Horizontal scroll

Design Tokens (Example — Replace With Your Own)

Note: The tokens below are examples to get you started. Replace them with your own design system values, or use the Figma MCP to extract tokens from your design files.

Colors

accent: #675DF4 (primary brand)
accent-hover: #746BF5
accent-pressed: #564BE7
surface-text: #1f2533 (default text)
surface-text-weak: #545f70 (secondary text)
negative: #A8242A (destructive/error)
positive: #00875a (success)

Spacing (4px Scale)

s1: 4px   s2: 8px   s3: 12px
s4: 16px  s5: 20px  s6: 24px  s8: 32px

Typography

text-headline-small: 20px/28px, SemiBold
text-title-medium: 16px/24px, SemiBold
text-title-small: 14px/20px, SemiBold
text-label-medium: 14px/20px, Medium
text-body-small: 14px/20px, Regular

Border Radius

rounded-xsmall: 4px
rounded-small: 8px (rounded-sm in template)
rounded-medium: 12px (rounded-md in template)
rounded-full: 9999px

Common Patterns

Tabs

<div class="flex gap-2 border-b border-surface-border-weak">
  <button class="tab-btn px-s4 py-s2 text-label-medium border-b-2 border-accent text-accent" data-tab="tab1">Tab 1</button>
  <button class="tab-btn px-s4 py-s2 text-label-medium border-b-2 border-transparent text-surface-text-weak hover:text-surface-text" data-tab="tab2">Tab 2</button>
</div>
<div id="tab1" class="tab-content py-s4">Content 1</div>
<div id="tab2" class="tab-content py-s4 hidden">Content 2</div>

<script>
document.querySelectorAll('.tab-btn').forEach(btn => {
  btn.addEventListener('click', () => {
    const target = btn.dataset.tab;
    document.querySelectorAll('.tab-content').forEach(t => t.classList.add('hidden'));
    document.getElementById(target).classList.remove('hidden');
    document.querySelectorAll('.tab-btn').forEach(b => {
      b.classList.remove('border-accent', 'text-accent');
      b.classList.add('border-transparent', 'text-surface-text-weak');
    });
    btn.classList.add('border-accent', 'text-accent');
    btn.classList.remove('border-transparent', 'text-surface-text-weak');
  });
});
</script>

Modal

<div id="my-modal" class="fixed inset-0 z-50 hidden bg-black/50">
  <div class="fixed inset-0 flex items-center justify-center p-4">
    <div class="bg-white rounded-md shadow-large p-s6 max-w-lg w-full">
      <h2 class="text-headline-small mb-s4">Modal Title</h2>
      <p class="text-body-small text-surface-text-weak mb-s6">Modal content here.</p>
      <div class="flex justify-end gap-s3">
        <button onclick="closeModal('my-modal')" class="focus-ring px-s4 py-s2 rounded-sm text-label-medium text-surface-text hover:bg-surface-hover">Cancel</button>
        <button class="focus-ring px-s4 py-s2 rounded-sm bg-accent text-white hover:bg-accent-hover text-label-medium">Confirm</button>
      </div>
    </div>
  </div>
</div>

<script>
function openModal(id) { document.getElementById(id).classList.remove('hidden'); }
function closeModal(id) { document.getElementById(id).classList.add('hidden'); }
</script>

Button Variants

<!-- Primary -->
<button class="focus-ring px-s4 py-s2 rounded-sm bg-accent text-white hover:bg-accent-hover text-label-medium">Primary</button>

<!-- Secondary -->
<button class="focus-ring px-s4 py-s2 rounded-sm border border-surface-border text-surface-text hover:bg-surface-hover text-label-medium">Secondary</button>

<!-- Tertiary -->
<button class="focus-ring px-s4 py-s2 rounded-sm text-accent hover:bg-accent-weak text-label-medium">Tertiary</button>

<!-- Destructive -->
<button class="focus-ring px-s4 py-s2 rounded-sm bg-negative text-white hover:bg-negative-hover text-label-medium">Delete</button>

Search Input

<div class="flex items-center gap-s2 h-10 w-80 rounded-sm border border-surface-border bg-white px-s3">
  <svg class="w-4 h-4 text-surface-text-weak" fill="none" viewBox="0 0 24 24" stroke="currentColor">
    <circle cx="11" cy="11" r="8"/><path d="m21 21-4.35-4.35"/>
  </svg>
  <input type="search" placeholder="Search..." class="flex-1 text-body-small outline-none" />
</div>

Output Checklist

Before delivering prototype, verify:

  • <title> updated to match prototype name
  • Main <h1> heading updated
  • All interactive elements have focus-ring class
  • Semantic HTML used (<nav>, <main>, <header>)
  • Skip link present at top of body
  • All SVG icons have aria-hidden="true"
  • JavaScript is vanilla (no frameworks)
  • File saved with descriptive name (e.g., user_management_prototype.html)

Figma Integration

If user provides a Figma URL, use Figma MCP tools:

mcp__figma__get_design_context - Extract design details
mcp__figma__get_screenshot - Get visual reference

Map Figma designs to available components from DesignSystem.html.


Example Invocations

Simple:

/rapid-prototype "Create a settings page with tabs for Profile, Security, and Notifications"

Detailed:

/rapid-prototype "Build a task management prototype with:
- Header with breadcrumbs (Home > Projects > My Tasks)
- Filter bar with status dropdown and search
- Task list with checkboxes, priority badges, due dates
- Each task row clickable to open detail modal
- Bulk action bar when tasks selected"

From Figma:

/rapid-prototype from https://www.figma.com/design/abc123/MyDesign?node-id=123-456

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.