Skip to content
cccskills
FreePublished on GitHub

css-development

This skill should be used when working with CSS, creating components, styling elements, refactoring styles, or reviewing CSS code. Triggers on "CSS", "styles", "Tailwind", "dark mode", "component styling", "semantic class", "@apply", "stylesheet". Routes to specialized sub-skills for creation, validation, or refactoring.

The English summary is being prepared. Showing the original description.

See how to install

Files included (5)

  • SKILL.md5.9 KB
  • create-component/SKILL.md10.7 KB
  • refactor/SKILL.md11.0 KB
  • skill-report.json183.4 KB
  • validate/SKILL.md9.7 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

CSS Development Skill

Overview

Comprehensive workflow for CSS development using Tailwind + semantic component patterns. This skill automatically routes you to the appropriate specialized workflow based on context.

This skill will invoke one of three sub-skills:

  • css-development:create-component - Creating new CSS components
  • css-development:validate - Reviewing existing CSS
  • css-development:refactor - Transforming CSS to semantic patterns

When This Skill Applies

Claude Code will automatically load this skill when you:

  • Create new CSS components or styles
  • Review or validate existing CSS code
  • Refactor inline styles or utility classes
  • Work with component styling in any framework
  • Need to add dark mode support
  • Write CSS tests

CSS Development Patterns

All sub-skills follow these core patterns. Reference this section when working with CSS.

Core Principles

  1. Semantic Naming - Use descriptive class names (.button-primary, .card-header) not utility names (.btn-blue, .card-hdr)
  2. Tailwind Composition - Leverage Tailwind utilities via @apply directive
  3. Dark Mode by Default - Include dark: variants for all colored/interactive elements
  4. Composition Over Creation - Reuse existing classes before creating new ones
  5. Test Coverage - Static CSS tests + component rendering tests
  6. Documentation - Usage comments above each component class

Component Class Pattern

/* Button component - Primary action button with hover states
   Usage: <button className="button-primary">Click me</button> */
.button-primary {
  @apply bg-indigo-500 hover:bg-indigo-700 dark:bg-indigo-600 dark:hover:bg-indigo-800;
  @apply px-6 py-3 rounded-lg font-medium text-white;
  @apply transition-all duration-200 hover:-translate-y-0.5;
}

Key characteristics:

  • Group related utilities logically (background, spacing, typography, transitions)
  • Include hover/focus/active states
  • Include dark mode variants using dark: prefix
  • Use Tailwind's built-in scales (indigo-500, gray-800, etc.)

File Structure Convention

styles/
├── components.css      # All semantic component classes
└── __tests__/
    └── components.test.ts  # CSS and component tests

Markup Integration (Framework-Agnostic)

Works with React, Vue, Svelte, or vanilla HTML:

React:

const classes = `button-primary ${className}`.trim();
<button className={classes}>...</button>

Vanilla HTML:

<button class="button-primary custom-class">...</button>

Vue:

<button :class="['button-primary', customClass]">...</button>

Key principle: Apply semantic class + allow additional classes for customization.

Atomic Design Levels

  • Atoms: Basic building blocks (.button, .input, .badge, .spinner)
  • Molecules: Composed components (.card, .form-field, .empty-state)
  • Organisms: Complex components (.page-layout, .session-card, .conversation-timeline)

Testing Pattern

Static CSS Tests:

it('should have button component classes', () => {
  const content = readFileSync('styles/components.css', 'utf-8');
  expect(content).toContain('.button-primary');
});

Component Rendering Tests:

it('applies semantic class and custom className', () => {
  render(<Button variant="primary" className="custom" />);
  expect(screen.getByRole('button')).toHaveClass('button-primary', 'custom');
});

Workflow: Context Detection and Routing

When this skill is invoked, follow these steps to route to the appropriate sub-skill:

Step 1: Analyze Context

Look at the user's request and recent conversation to determine intent:

Creating new components?

  • Keywords: "create", "add", "new component", "build a", "make a"
  • Files: Mention of components.css or new component names
  • Intent: User wants to add new CSS

Validating existing CSS?

  • Keywords: "review", "validate", "check", "audit", "look at"
  • Files: Reference to existing CSS files or components
  • Intent: User wants feedback on existing CSS

Refactoring CSS?

  • Keywords: "refactor", "clean up", "extract", "improve", "convert"
  • Code: Inline styles or utility classes in markup visible
  • Intent: User wants to transform existing CSS patterns

Step 2: Choose Sub-Skill

Based on context analysis:

If creating: Use the Skill tool to invoke css-development:create-component

If validating: Use the Skill tool to invoke css-development:validate

If refactoring: Use the Skill tool to invoke css-development:refactor

If ambiguous: Ask the user using AskUserQuestion tool:

Question: "What would you like to do with CSS?"
Options:
  - "Create new component" (Guide creating new semantic CSS component classes)
  - "Validate existing CSS" (Review CSS against established patterns)
  - "Refactor CSS" (Transform inline/utility styles to semantic components)

Step 3: Invoke Sub-Skill

Use the Skill tool to invoke the chosen sub-skill:

Example:

I'm routing you to the create-component workflow.
[Invoke Skill tool with skill: "css-development:create-component"]

Step 4: Hand Off Control

Once the sub-skill is invoked, it takes over. The main skill's job is complete.

Important Notes

  • Don't skip routing: Always analyze context and choose the right sub-skill
  • Don't duplicate sub-skill logic: Let sub-skills handle their workflows
  • Reference pattern documentation: Sub-skills will reference the patterns documented above
  • User can invoke directly: User can call sub-skills directly (e.g., "use css-development:validate")

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Trigger for: explaining current FAR, DFARS, or agency-supplement text; determining documented acquisition-policy status for an agency and FAR part; comparing codified text, RFO model text, and agency deviations; tracing acquisition rulemaking; finding procurement comment periods; analyzing public comments; refreshing prior policy analysis; or producing an Acquisition Policy Impact Brief. Use published federal sources and supplied policy documents without making legal or procurement-specific applicability determinations. Do not use for clause selection, market research execution, opportunities, pricing, grants, or cooperative agreements.

The English summary is being prepared. Showing the original description.

aiskillstore/marketplace439Updated Oct 19, 2026

Systematic framework for extracting actionable items from documents and organizing them into prioritized, trackable checklists. Use when converting reports, meeting notes, audits, or any document with embedded action items into structured TODO lists.

The English summary is being prepared. Showing the original description.

aiskillstore/marketplace439Updated Oct 19, 2026

Browser automation for AI agents via inference.sh. Navigate web pages, interact with elements using @e refs, take screenshots, record video. Capabilities: web scraping, form filling, clicking, typing, drag-drop, file upload, JavaScript execution. Use for: web automation, data extraction, testing, agent browsing, research. Triggers: browser, web automation, scrape, navigate, click, fill form, screenshot, browse web, playwright, headless browser, web agent, surf internet, record video

The English summary is being prepared. Showing the original description.

aiskillstore/marketplace439Updated Oct 19, 2026

This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.

The English summary is being prepared. Showing the original description.

aiskillstore/marketplace439Updated Oct 19, 2026

Run AI apps via inference.sh CLI - image generation, video creation, LLMs, search, 3D, Twitter automation. Models: FLUX, Veo, Gemini, Grok, Claude, Seedance, OmniHuman, Tavily, Exa, OpenRouter, and many more. Use when running AI apps, generating images/videos, calling LLMs, web search, or automating Twitter. Triggers: inference.sh, infsh, ai model, run ai, serverless ai, ai api, flux, veo, claude api, image generation, video generation, openrouter, tavily, exa search, twitter api, grok

The English summary is being prepared. Showing the original description.

aiskillstore/marketplace439Updated Oct 19, 2026

Agile ceremonies and sprint management specialist. Use when running standups, retrospectives, sprint planning, or tracking sprint progress.

The English summary is being prepared. Showing the original description.

aiskillstore/marketplace439Updated Oct 19, 2026

See all skills by aiskillstore

Report a problem with this skill