Back to Skills

Design Review

7-phase frontend design review with accessibility (WCAG 2.1 AA), responsive testing, visual polish. Use for PR reviews, UI audits, or encountering contrast issues, broken layouts, accessibility violations, inconsistent spacing, missing focus states.

testing
By secondsky
18028Updated 1 week agoTypeScriptMIT

Skill Content

# Design Review Skill

**Status**: Production Ready ✅
**Last Updated**: 2025-11-20
**Dependencies**: Playwright MCP or Chrome DevTools
**Methodology**: 7-phase systematic review (inspired by Stripe, Airbnb, Linear)

---

## Quick Start

### 1. Prerequisites Check

Before starting a design review, verify browser automation tools are available:

**Option A: Playwright MCP** (recommended for interactive testing)
- See the `playwright-testing` skill for Playwright setup
- Provides browser automation, screenshots, viewport testing, console monitoring

**Option B: Chrome DevTools CLI** (alternative for screenshots and performance)
- See the `chrome-devtools` skill for Puppeteer CLI setup
- Provides screenshot capture, performance analysis, network monitoring

For complete browser tools reference, see [references/browser-tools-reference.md](references/browser-tools-reference.md).

### 2. Understand the Review Scope

**For PR reviews:**
```bash
# Analyze git diff to understand scope
git diff --name-only origin/main...HEAD

# Read PR description for context
```

**For general UI reviews:**
Simply provide the preview URL and component/page description.

### 3. Execute 7-Phase Review

Follow the systematic checklist below. Each phase has specific objectives and testing procedures.

---

## The 7-Phase Review Methodology

### Phase 0: Preparation

**Objective:** Understand context and set up testing environment.

**Steps:**
1. **Read PR description** or review request to understand:
   - Motivation for changes
   - Scope of implementation
   - Testing notes from developer
   - Expected behavior

2. **Analyze code diff** (if PR available):
   ```bash
   git diff origin/main...HEAD
   ```
   Identify modified files (components, styles, tests)

3. **Set up live preview environment:**
   - Navigate to preview URL using browser tools
   - Set initial viewport: 1440x900 (desktop)
   - Take baseline screenshot for reference

4. **Review design principles** (if project has custom guidelines):
   - Check project CLAUDE.md for design standards
   - Review component library documentation
   - Note design system tokens and patterns

**When to skip:** For quick component reviews without git context.

---

### Phase 1: Interaction & User Flow

**Objective:** Verify the interactive experience works as expected.

**For complete interaction guide**: Load `references/interaction-patterns.md` when testing interactive states, forms, buttons, navigation flows, micro-interactions, modals, or keyboard navigation.

**Quick checklist:**
- Test 5 interactive states (default, hover, active, focus, disabled) for all elements
- Execute primary user flow (form submission, navigation, key actions)
- Verify destructive actions have confirmation dialogs
- Assess perceived performance (loading states, optimistic UI)

**Triage:** [Blocker] Critical flow broken | [High] Poor UX/missing focus states | [Medium] Missing polish | [Nitpick] Minor timing issues

---

### Phase 2: Responsiveness Testing

**Objective:** Ensure design works across all viewport sizes.

**For complete responsive guide**: Load `references/responsive-testing.md` when testing viewports, touch targets, mobile navigation, image responsiveness, or debugging horizontal scrolling.

**Test 3 viewports:**
- **Desktop (1440px)**: Optimal layout, full feature set
- **Tablet (768px)**: Graceful adaptation, 44×44px touch targets, collapsing nav
- **Mobile (375px)**: No horizontal scroll, 16px min text, mobile-friendly navigation

**Quick testing:**
```bash
mcp__playwright__browser_resize(width: 1440, height: 900)  # Desktop
mcp__playwright__browser_resize(width: 768, height: 1024)  # Tablet
mcp__playwright__browser_resize(width: 375, height: 667)   # Mobile
mcp__playwright__browser_take_screenshot(fullPage: true)
```

**Triage:** [Blocker] Layout broken | [High] Horizontal scroll/overlapping | [Medium] Suboptimal spacing | [Nitpick] Minor inconsistencies

---

### Phase 3: Visual Polish

**Objective:** Assess aesthetic quality and visual consistency.

**For design principles guide**: Load `references/visual-polish.md` when evaluating typography hierarchy, spacing/layout, color palette, alignment/grid, visual hierarchy, image quality, or S-Tier design standards.

**Quick evaluation (5 criteria):**
1. **Layout & spacing**: Grid alignment, 8px scale, design tokens (no magic numbers like 17px)
2. **Typography**: Clear H1>H2>H3 hierarchy, 1.5-1.7 line height, limited font weights
3. **Color**: Design system tokens, semantic usage (red=error, green=success), consistent brand
4. **Images**: High-res (no pixelation), correct aspect ratios, optimized sizes, alt text
5. **Visual hierarchy**: Primary actions stand out, eye flows naturally, strategic whitespace

**Triage:** [Blocker] Illegible text/broken images | [High] Obvious inconsistencies | [Medium] Spacing/alignment issues | [Nitpick] Aesthetic preferences

---

### Phase 4: Accessibility (WCAG 2.1 AA)

**Objective:** Ensure inclusive design for all users.

**For complete WCAG 2.1 AA checklist**: Load `references/accessibility-wcag.md` when verifying WCAG compliance, testing keyboard navigation, checking color contrast, auditing semantic HTML, or using accessibility testing tools (Lighthouse, axe, WAVE).

**Quick WCAG tests (4 principles):**

1. **Perceivable**: Alt text on images, color contrast (4.5:1 text, 3:1 UI components), semantic HTML
2. **Operable**: Keyboard navigation (Tab order logical, visible focus on ALL interactive elements, Enter/Space activation, Escape closes modals, no keyboard traps)
3. **Understandable**: Clear labels, helpful error messages, consistent navigation/terminology
4. **Robust**: Valid HTML, proper ARIA attributes (roles, states, properties)

**Critical tests:**
- Tab through entire page (verify focus states visible, logical order, no traps)
- Test with WebAIM Contrast Checker (all text/UI ≥4.5:1 or 3:1)
- Verify form labels associated with inputs (`<label for="id">` or `aria-label`)
- Check semantic HTML (h1→h2→h3 no skipping, `<button>` not `<div onClick>`)

**Triage:** [Blocker] No keyboard access to core features | [High] WCAG AA violations | [Medium] Semantic HTML issues | [Nitpick] Enhanced accessibility

---

### Phase 5: Robustness Testing

**Objective:** Verify handling of edge cases and error conditions.

**Test scenarios:**

#### 5.1 Form Validation

- Submit form with empty required fields
- Enter invalid data (wrong email format, out-of-range numbers)
- Test field-level validation (real-time feedback)
- Verify clear error messages with guidance
- Test successful submission flow (confirmation message)

#### 5.2 Content Overflow

- **Long text strings**: Very long names, emails, titles
- **Many items**: Large lists, tables with hundreds of rows
- **Deeply nested content**: Comments with many replies
- **Empty states**: No data to display (show helpful message)

**Common overflow issues:**
- Text breaking layout (overflowing containers)
- Truncation without ellipsis or tooltip
- Performance issues with large lists
- Missing empty state designs

#### 5.3 Loading & Error States

- **Loading states**: Skeleton screens, spinners, progress indicators
- **Error messages**: Clear, actionable error descriptions
- **Retry mechanisms**: Allow user to retry failed operations
- **Timeout handling**: Graceful handling of slow/failed requests
- **Optimistic updates**: Immediate feedback, rollback on failure

**Test procedure:**
```bash
# Simulate slow network
# Check browser DevTools Network tab → throttling

# Force error states
# Test with invalid API responses or network failures
```

**Common problems:**
- No loading indicators (appears frozen)
- Vague error messages ("Error occurred")
- No retry mechanism after failures
- Layout jumps when content loads

**Triage priorities:**
- **[Blocker]** Crashes or complete failures under edge cases
- **[High]** Poor error handling or confusing states
- **[Medium]** Missing edge case handling or minor issues
- **[Nitpick]** Loading state aesthetics or minor polish

---

### Phase 6: Code Health

**Objective:** Ensure maintainable, consistent implementation.

**For code patterns guide**: Load `references/code-health-patterns.md` when evaluating component reuse (DRY principle), design token usage (colors, spacing, typography), pattern consistency (naming, file structure, API patterns), or identifying red flags (duplication, magic numbers, broken abstractions).

**Quick review (3 criteria):**
1. **Component reuse**: No copy-paste, shared components extracted, composition over duplication
2. **Design tokens**: CSS variables for colors/spacing/typography (no magic numbers like `margin: 17px`), border radii consistent
3. **Pattern consistency**: Follows codebase patterns, naming conventions match, file structure organized

**Triage:** [High] Introduces tech debt/breaks patterns | [Medium] Missed reuse opportunities | [Nitpick] Code style preferences

---

### Phase 7: Content & Console

**Objective:** Verify polished details and technical correctness.

#### 7.1 Content Review

**Check for:**
- **Grammar and spelling**: No typos or grammatical errors
- **Clarity**: Labels and instructions are unambiguous
- **Tone consistency**: Matches brand voice (formal/casual)
- **Placeholder text**: Replaced with real content (no "Lorem ipsum")
- **Microcopy quality**: Helpful error messages, button labels, tooltips

**Common content issues:**
- Typos in UI text
- Placeholder text left in production
- Vague labels ("Submit" vs "Save Changes")
- Inconsistent terminology
- Unhelpful error messages ("Error" vs "Email format invalid")

#### 7.2 Console Check

**Test procedure:**
```bash
# Using Playwright MCP
mcp__playwright__browser_console_messages()

# Using Chrome DevTools
# Open DevTools → Console tab
```

**Look for:**
- **JavaScript errors**: Uncaught exceptions, null references
- **React warnings**: Key prop warnings, lifecycle issues
- **Network failures**: Failed API requests, 404s
- **Deprecation warnings**: Old API usage warnings
- **Performance warnings**: Slow renders, memory leaks

**Triage priorities:**
- **[Blocker]** Console errors breaking functionality
- **[High]** Grammar errors or confusing content in user-facing text
- **[Medium]** Console warnings or minor content issues
- **[Nitpick]** Content polish, minor console noise

---

## Communication Principles

### 1. Problems Over Prescriptions

Describe the **problem and its impact**, not the solution. Let the developer decide implementation.

**❌ Prescriptive (avoid):**
"Change the margin to 16px"

**✅ Problem-focused (preferred):**
"The spacing feels inconsistent with adjacent elements, creating visual clutter that distracts from the primary CTA. The current spacing breaks the established rhythm of the design system."

### 2. Triage Matrix

Categorize **every issue** with clear priority:

| Priority | Criteria | Action Required |
|----------|----------|----------------|
| **[Blocker]** | Critical failures, core functionality broken, critical accessibility violations | Must fix before merge |
| **[High-Priority]** | Significant UX issues, obvious design inconsistencies, WCAG violations | Should fix before merge |
| **[Medium-Priority]** | Improvements, minor inconsistencies, edge case handling | Consider for follow-up PR |
| **[Nitpick]** | Aesthetic preferences, minor polish, subjective opinions | Optional refinements |

**Important:** Prefix all nitpicks with "Nit:" to signal low priority.

### 3. Evidence-Based Feedback

Always provide **screenshots** for visual issues. Screenshots should:
- Show the problem clearly
- Include relevant context (surrounding elements)
- Indicate what to look at (arrows, highlights if needed)

**Example:**
```markdown
### [High-Priority] Poor contrast on disabled button

**Problem:** Disabled button text has insufficient contrast (2.1:1), failing WCAG AA
standard (requires 4.5:1). Users with low vision may not recognize the button as disabled.

**Screenshot:** [Attach screenshot showing disabled button]

**Impact:** Accessibility violation, potential confusion for users with visual impairments.
```

### 4. Start with Positives

Always acknowledge what works well before listing issues. This:
- Shows you recognize good work
- Provides balanced feedback
- Maintains positive collaboration

**Example:**
```markdown
### Design Review Summary

The new checkout flow shows excellent attention to user experience. The step indicator
is clear and well-designed, error messages are helpful and actionable, and the overall
layout feels spacious and uncluttered. The loading states with skeleton screens are
particularly well-executed. Great work on the form validation feedback!

However, there are a few accessibility and responsiveness issues to address before merge...
```

---

## Report Structure Template

**For complete template**: Load `assets/review-report-template.md` for the full markdown template with all sections and examples.

**Essential structure:**

```markdown
## Design Review Summary
[2-3 sentences: positive acknowledgment + overall assessment]
**Review scope:** [PR #, pages, components]
**Viewports tested:** Desktop (1440px), Tablet (768px), Mobile (375px)
**Methodology:** 7-phase comprehensive review

---

### Findings

#### 🚨 Blockers
[Critical issues requiring immediate fix before merge]
- **[Blocker] [Title]**: Problem + Screenshot + Phase

#### ⚠️ High-Priority Issues
[Significant issues to fix before merge]
- **[High] [Title]**: Problem + Screenshot + Phase

#### 📋 Medium-Priority / Suggestions
[Improvements for follow-up PR]
- **[Medium] [Title]**: Problem + Phase

#### ✨ Nitpicks
[Minor aesthetic details - optional]
- **Nit:** [Issue] - [Brief description]

---

### Testing Evidence
**Screenshots:** Desktop (1440px) + Tablet (768px) + Mobile (375px)
**Console output:** [Errors/warnings or "Console clean"]
**Accessibility:** Keyboard nav + Focus states + Color contrast

---

### Next Steps
1. Fix Blockers
2. Address High-Priority issues
3. Consider Medium-Priority items

**Overall assessment:** [Ready to merge after blockers fixed / Needs revisions / Ready to merge!]
```

---

## When to Load References

Load reference files when working on specific aspects of design review:

### accessibility-wcag.md
Load when:
- **Standards-based**: Verifying WCAG 2.1 AA compliance for production deployment
- **Issue-based**: Encountering accessibility violations (color contrast, keyboard navigation, semantic HTML, focus states, ARIA attributes)
- **Testing-based**: Conducting comprehensive accessibility audit with systematic checklist
- **Tools-based**: Using accessibility testing tools (Lighthouse, axe, WAVE, Pa11y) for automated testing
- **Triage-based**: Determining severity of accessibility issues (Blocker/High/Medium for WCAG violations)

### browser-tools-reference.md
Load when:
- **Setup-based**: Installing or configuring Playwright MCP or Chrome DevTools CLI for testing
- **Command-based**: Need specific Playwright commands (navigate, resize viewport, screenshot, click, type, hover, get console output)
- **Workflow-based**: Implementing common testing workflows (responsive review across 3 viewports, form interaction testing, keyboard navigation testing)
- **Selector-based**: Struggling with CSS selectors, text selectors, or accessibility selectors for element targeting
- **Troubleshooting-based**: Playwright MCP not finding elements, Chrome dependencies missing, screenshot capture issues

### code-health-patterns.md
Load when:
- **Pattern-based**: Evaluating component reuse patterns, DRY principle compliance, extracting shared components
- **Token-based**: Checking design token usage (colors, spacing scale, typography scale, border radii consistency)
- **Consistency-based**: Reviewing pattern consistency (naming conventions, file structure organization, API patterns, state management)
- **Example-based**: Need code examples comparing good vs bad patterns (inline styles vs tokens, duplication vs composition)
- **Red-flag-based**: Identifying code health issues (copy-paste duplication, magic numbers like `17px`, inconsistent state management, broken abstractions)

### design-principles-s-tier.md
Load when:
- **Standards-based**: Ensuring S-Tier SaaS dashboard quality (Stripe, Airbnb, Linear, Vercel level polish)
- **System-based**: Evaluating design system foundation (color palette structure, typography scale, spacing scale, core UI components)
- **Module-based**: Reviewing specific modules (multimedia moderation interfaces, data tables, configuration panels, dashboards)
- **Philosophy-based**: Applying core design philosophy (users first, meticulous craft over speed, simplicity over complexity, consistency)
- **Architecture-based**: Evaluating CSS & styling architecture (design tokens, component patterns, responsive strategies)

### interaction-patterns.md
Load when:
- **States-based**: Testing interactive states (default, hover, active, focus, disabled) for buttons, inputs, links
- **Form-based**: Testing form interactions, validation patterns, error states, success states, loading states
- **Button-based**: Evaluating button loading states, destructive action confirmation patterns, primary vs secondary actions
- **Flow-based**: Testing navigation flows, user journeys, multi-step processes, modal interactions
- **Animation-based**: Reviewing micro-interactions, animation timing (200-300ms), perceived performance (optimistic UI, skeleton screens)
- **Modal-based**: Testing modal interactions, keyboard traps, focus management, Escape key behavior

### responsive-testing.md
Load when:
- **Viewport-based**: Testing at specific viewports (desktop 1440px, tablet 768px, mobile 375px) with Playwright MCP
- **Touch-based**: Verifying touch target sizes meet minimum 44×44px requirement for mobile usability
- **Overflow-based**: Debugging horizontal scrolling issues or layout overflow problems on mobile
- **Mobile-based**: Ensuring text readability (16px minimum font size), mobile navigation patterns, responsive images
- **Breakpoint-based**: Implementing or testing breakpoint strategy (common breakpoints: 640px, 768px, 1024px, 1280px)
- **Navigation-based**: Testing responsive navigation patterns (hamburger menus, collapsing navigation, mobile drawer menus)

### visual-polish.md
Load when:
- **Typography-based**: Evaluating font hierarchy (H1>H2>H3), font scale standards (16/18/24/32/48/64px), line height (1.5-1.7), readability
- **Spacing-based**: Checking 8-point grid compliance (8/16/24/32/40/48/64px), consistent spacing scale, component padding/margin
- **Color-based**: Verifying color palette consistency, semantic color usage (red=error, green=success), design token usage (no hardcoded hex values)
- **Alignment-based**: Checking grid-based layout, precise alignment (0.5px precision), vertical rhythm, visual balance
- **Hierarchy-based**: Evaluating visual hierarchy techniques (size contrast, weight contrast, color contrast, position, strategic whitespace)
- **Quality-based**: Assessing image quality (no pixelation), correct aspect ratios, proper image optimization for web
- **Component-based**: Reviewing design system components (button styles, form input styles, card components, consistent border radii)

---

## Known Issues Prevention

This skill prevents **8** documented design review issues:

| Issue | Problem | Impact | Prevention |
|-------|---------|--------|------------|
| **#1: Missing Accessibility** | Reviews focus only on visual appearance, ignoring keyboard navigation and screen readers | WCAG violations shipped to production, excluding users with disabilities | Phase 4 enforces complete WCAG 2.1 AA checklist with keyboard testing |
| **#2: Incomplete Responsive Testing** | Reviewing only at desktop viewport, missing mobile breakage | Broken mobile layouts, frustrated mobile users | Phase 2 requires testing at 1440px, 768px, and 375px viewports |
| **#3: Vague Feedback** | Comments like "looks off" without screenshots or specifics | Wasted time, unclear action items, frustrated developers | Evidence-based feedback principle requires screenshots |
| **#4: Prescriptive Solutions** | Dictating implementation ("change margin to 16px") instead of describing UX impact | Design-dev friction, missed better solutions | "Problems Over Prescriptions" principle enforced |
| **#5: No Triage Priority** | All feedback treated equally, blocking merges on nitpicks | Slowed delivery, unclear priorities | Triage matrix (Blocker/High/Medium/Nitpick) required |
| **#6: Skipped Edge Cases** | Happy path works, but error states and overflow break layout | Production bugs with edge cases | Phase 5 mandates robustness testing |
| **#7: Console Errors Ignored** | Visual design passes, but JavaScript errors exist in console | Runtime failures, poor user experience | Phase 7 requires console check |
| **#8: Inconsistent Methodology** | Ad-hoc reviews miss critical areas depending on reviewer mood | Incomplete reviews, missed issues | 7-phase checklist ensures comprehensive, repeatable reviews |

---

## Dependencies

### Required

**Browser automation tools** (one of the following):

1. **Playwright MCP** (recommended)
   - See `playwright-testing` skill for installation
   - Provides: Browser automation, screenshots, viewport testing, console monitoring
   - Best for: Interactive testing, keyboard navigation, form testing

2. **Chrome DevTools CLI**
   - See `chrome-devtools` skill for installation
   - Provides: Screenshot capture, performance analysis, network monitoring
   - Best for: Visual testing, performance audits

**Live preview environment:**
- URL accessible for testing
- Represents actual implementation (not mockups)

### Optional

- **Git/GitHub**: For PR context and diff analysis
- **Design system docs**: For consistency checks against established patterns
- **Project CLAUDE.md**: For project-specific design guidelines

### Installation Guidance

If browser tools are not available, this skill will:
1. Detect missing tools
2. Link to appropriate skill for installation (`playwright-testing` or `chrome-devtools`)
3. Provide fallback guidance for manual testing

---

## Related Skills

- **playwright-testing**: E2E testing with Playwright, browser automation setup
- **chrome-devtools**: Browser automation via Puppeteer CLI scripts
- **frontend-design**: Create new frontend interfaces with design quality (complementary skill)
- **tailwind-v4-shadcn**: UI framework implementation (designs being reviewed may use this)
- **ai-sdk-ui**: AI-powered UI components (may be part of reviewed interfaces)

---

## Official Documentation

- **WCAG 2.1 Guidelines**: https://www.w3.org/WAI/WCAG21/quickref/
- **WebAIM Contrast Checker**: https://webaim.org/resources/contrastchecker/
- **Playwright Documentation**: https://playwright.dev/
- **Inclusive Design Principles**: https://inclusivedesignprinciples.org/
- **A11y Project Checklist**: https://www.a11yproject.com/checklist/

---

## Production Validation

**This skill is based on real design review workflows** used at:
- **Methodology inspiration**: Stripe, Airbnb, Linear (7-phase systematic approach)
- **Testing approach**: Automated browser testing with Playwright/Puppeteer
- **Accessibility standards**: WCAG 2.1 AA compliance (industry standard)

**Estimated token efficiency:**
- Without skill: ~25k tokens (trial-and-error, repeated corrections)
- With skill: ~8k tokens (guided methodology, systematic approach)
- **Savings: ~68%** with 100% checklist coverage

---

**Questions or issues?**

1. Check [references/accessibility-wcag.md](references/accessibility-wcag.md) for complete WCAG checklist
2. See [references/browser-tools-reference.md](references/browser-tools-reference.md) for Playwright/Chrome DevTools commands
3. Review [references/visual-polish.md](references/visual-polish.md) for design principles
4. Verify browser tools are installed (see `playwright-testing` or `chrome-devtools` skills)
5. Ensure preview URL is live and accessible

How to use

  1. Copy the skill content above
  2. Create a .claude/skills directory in your project
  3. Save as .claude/skills/claude-skills-design-review.md
  4. Use /claude-skills-design-review in Claude Code to invoke this skill

Claude Code Skills Collection

170 production-ready skills for Claude Code CLI

Version 3.3.1 | Last Updated: 2026-05-14

<div align="center">

🔌 Platform Support

This repository uses Claude Plugin Patterns — natively supported by:

PlatformStatusNotes
Claude CodeNativeFull marketplace support
Factory DroidNativeFull marketplace support
</div> **For all other Platforms like opencode, codex and others, you can use https://github.com/enulus/OpenPackage **

A curated collection of battle-tested skills for building modern web applications with Cloudflare, AI integrations, React, Tailwind, and more.

PS: if skills.sh warns about any skill: Their scan process is a outdated LLM which flags newest versions pins (like in ZOD) as non existent and by that potentially malicous.


Quick Start

Marketplace Installation (Recommended)

# Add the marketplace
/plugin marketplace add https://github.com/secondsky/claude-skills

# Install individual skills as needed
/plugin install cloudflare-d1@claude-skills
/plugin install tailwind-v4-shadcn@claude-skills
/plugin install ai-sdk-core@claude-skills

See MARKETPLACE.md for complete catalog of all 170 skills.

Bulk Installation (Contributors)

# Clone the repository
git clone https://github.com/secondsky/claude-skills.git
cd claude-skills

# Install all 170 skills at once
./scripts/install-all.sh

# Or install individual skills
./scripts/install-skill.sh cloudflare-d1

Repository Structure

This repository contains 170 production-tested skills for Claude Code, each focused on a specific technology or capability.

Individual Skills: Each skill is a standalone unit with:

  • SKILL.md - Core knowledge and guidance
  • Templates - Working code examples
  • References - Extended documentation
  • Scripts - Helper utilities

Installation Options:

  1. Individual - Install only the skills you need via marketplace
  2. Bulk - Install all 170 skills using ./scripts/install-all.sh

Available Skills (170 Individual Skills)

Each skill is individually installable. Install only the skills you need.

Full Catalog: See MARKETPLACE.md for detailed listings.

Categories

CategorySkillsExamples
tooling29turborepo, plan-interview, code-review
frontend26nuxt-v4, nuxt-v5, tailwind-v4-shadcn, tanstack-query, nuxt-studio, maz-ui, threejs
cloudflare21cloudflare-d1, cloudflare-workers-ai, cloudflare-agents
ai20openai-agents, claude-api, ai-sdk-core
api16api-design-principles, graphql-implementation
web10hono-routing, firecrawl-scraper, web-performance
mobile7swift-best-practices, react-native-app, react-native-skills
database6drizzle-orm-d1, neon-vercel-postgres, supabase-postgres-best-practices
security6csrf-protection, access-control-rbac
auth4better-auth
testing4vitest-testing, playwright-testing
design4design-review, design-system-creation
woocommerce4woocommerce-backend-dev
cms4hugo, sveltia-cms, wordpress-plugin-core
architecture3microservices-patterns, architecture-patterns
data3sql-query-optimization, recommendation-engine
seo2seo-optimizer, seo-keyword-cluster-builder
documentation1technical-specification

How It Works

Auto-Discovery

Claude Code automatically checks ~/.claude/skills/ for relevant skills before planning tasks:

User: "Set up a Cloudflare Worker with D1 database"
           ↓
Claude: [Checks skills automatically]
           ↓
Claude: "Found cloudflare-d1 skills.
         These prevent 12 documented errors. Use them?"
           ↓
User: "Yes"
           ↓
Result: Production-ready setup, zero errors, ~65% token savings

Note: Due to token limits, not all skills may be visible at once. See ⚠️ Important: Token Limits below.

Skill Structure

Each skill includes:

skills/[skill-name]/
├── SKILL.md              # Complete documentation
├── .claude-plugin/
│   └── plugin.json       # Plugin metadata
├── templates/            # Ready-to-copy templates
├── scripts/              # Automation scripts
└── references/           # Extended documentation

Recent Additions

May 2026

Supply Chain Security (cross-cutting):

  • dependency-upgrade expanded with Socket CLI integration — proactive malicious package detection, typosquatting alerts, and CI/CD security gates. New 418-line reference guide, 2 GitHub Actions templates, and expanded supply chain security comparison (3 tools)
  • 31 skills now include "Secure Installation" guidance — contextually-tailored security sections across all high-risk skill categories (scaffolding, MCP/agent SDKs, multi-provider installs, Docker, CI/CD). Covers 8 Bun skills, 5 Nuxt skills, 6 Cloudflare skills, 4 AI/agent skills, and 8 frontend/tooling skills
  • Supply chain security is now a first-class cross-cutting concern woven into the skill collection — not a standalone topic

February - April 2026

Full-Stack Frameworks:

  • nuxt-v5 (v1.0.0) - Full Nuxt 5 support with 4 skills (core, data, server, production), 3 diagnostic agents, and interactive setup wizard
  • supabase-postgres-best-practices - 30 Postgres optimization rules from Supabase across 8 categories
  • threejs (v1.0.0) - 3D web graphics: scenes, geometries, shaders, animations, post-processing

Infrastructure:

  • JSON schema validation - Automated plugin.json validation with CI support
  • GitHub issue templates - Skill-specific issue templates for bug reports, feature requests, and submissions

Plugin Enhancements:

  • mutation-testing - Added Bun native runner support
  • dependency-upgrade - Added supply chain security content

December 2025 - January 2026

Frontend Expansion:

  • nuxt-studio (v1.0.0) - Visual CMS for Nuxt Content with live preview, OAuth auth, and R2 storage integration
  • maz-ui (v1.0.0) - 50+ Vue/Nuxt components with theming, i18n, form generation, and 14 composables

Developer Workflow:

  • plan-interview (v2.0.0) - Adaptive interview-driven spec generation with autonomous quality review
  • turborepo (v2.8.0) - Updated to official Vercel skill with enhanced monorepo build optimization

Mobile Development:

  • react-native-skills (v1.0.0) - React Native & Expo best practices with performance optimization patterns

Enhanced Authentication:

  • better-auth (v2.2.0) - Expanded to 18 framework integrations with 30+ authentication plugins

⚠️ Important: Token Limits

Skill Visibility Constraint

Claude Code has a 15,000 character limit for the total size of skill descriptions in the system prompt. This limit also applies to commands and agents.

What this means:

  • Not all 170 skills may be visible in Claude's context at once
  • Skills are loaded based on relevance and available token budget
  • You can verify how many skills Claude currently sees by asking: "How many skills do you see in your system prompt?"

Checking Visible Skills

To verify which skills are currently loaded:

# Ask Claude Code directly
"Check what skills/plugins you see in your system prompt"

Claude will report something like: "85 of 170 skills visible due to token limits"

Workaround: Increase Token Budget

You can double the headroom for skill descriptions by setting an environment variable:

# Increase limit to 30,000 characters
export SLASH_COMMAND_TOOL_CHAR_BUDGET=30000

# Then launch Claude Code
claude

This gives you approximately 2x more skill visibility in the system prompt.

Note: This is a temporary workaround. The Claude Code team is working on better solutions for skill discovery and loading.


Token Efficiency

MetricManual SetupWith SkillsSavings
Average Tokens12,000-15,0004,000-5,000~65%
Typical Errors2-4 per service0 (prevented)100%
Setup Time2-4 hours15-45 minutes~80%

Across all 170 skills: 400+ documented errors prevented.


Contributing

Prerequisites for Contributors

Install the official plugin development toolkit:

/plugin install plugin-dev@claude-code-marketplace

This provides:

  • /plugin-dev:create-plugin command (8-phase guided workflow)
  • 7 comprehensive skills (hooks, MCP, structure, agents, commands, skills)
  • 2 specialized agents (agent-creator, plugin-validator)

Quick Steps

  1. Create skill directory in plugins/
  2. Add SKILL.md with YAML frontmatter
  3. Run ./scripts/sync-plugins.sh
  4. Submit pull request

See CONTRIBUTING.md and PLUGIN_DEV_BEST_PRACTICES.md for detailed guidelines.


Documentation

DocumentPurpose
START_HERE.mdStart here! Quick navigation guide
PLUGIN_DEV_BEST_PRACTICES.mdRepository-specific best practices (marketplace, budget, quality)
MARKETPLACE.mdFull skill catalog and installation guide
MARKETPLACE_MANAGEMENT.mdTechnical infrastructure (plugin.json, scripts, validation)
CLAUDE.mdProject context and development standards
CONTRIBUTING.mdContribution guidelines

Links


Built with ❤️ by Claude Skills Maintainers

View source on GitHub