@neyugn/agent-kits 0.5.0 → 0.5.3

package/kits/coder/skills/docker-patterns/SKILL.md

@@ -8,19 +8,26 @@ priority: HIGH
 
 # Docker Patterns - Container Excellence
 
-> **Philosophy:** Containers should be **small, secure, and consistent** across all environments. Optimize for production from day one.
+## ⚡ Quick Reference
+
+- **Multi-stage builds**: Separate builder + runtime · Copy only compiled artifacts · Never include src in prod image
+- **Base image**: `node:20-alpine` not `node:20` · Pin versions explicitly · Never use `latest`
+- **Security**: Non-root user (`USER node`) · Read-only filesystem · No secrets in image layers
+- **Layers**: `COPY package.json` before `COPY src` · Dependencies cached separately from code
+- **Compose**: Always `healthcheck` · Named volumes not bind mounts for data · `--wait` in CI
+- **Image size**: `apk add --no-cache` · `.dockerignore` always · target < 200MB for Node apps
+
+---
 
 ---
 
 ## Core Principles
 
-| Principle        | Rule                                                  |
-| ---------------- | ----------------------------------------------------- |
-| **Immutable**    | Build once, run anywhere - no runtime modifications   |
-| **Minimal**      | Only include what's needed - smaller = faster + safer |
-| **Secure**       | Non-root by default, no secrets in layers             |
-| **Reproducible** | Pin versions, multi-stage for consistent builds       |
-| **Observable**   | Health checks, structured logs, metrics endpoints     |
+- **Immutable**: Build once, run anywhere - no runtime modifications
+- **Minimal**: Only include what's needed - smaller = faster + safer
+- **Secure**: Non-root by default, no secrets in layers
+- **Reproducible**: Pin versions, multi-stage for consistent builds
+- **Observable**: Health checks, structured logs, metrics endpoints
 
 ---
 
@@ -72,26 +79,22 @@ CMD ["node", "dist/index.js"]
 
 ## Layer Optimization Rules
 
-| Rule                          | Why                                      |
-| ----------------------------- | ---------------------------------------- |
-| **COPY package.json first**   | Deps layer cached until deps change      |
-| **Combine RUN commands**      | Fewer layers = smaller image             |
-| **Clean in same RUN**         | `npm ci && npm cache clean` not separate |
-| **Order by change frequency** | Least changing → Most changing           |
-| **.dockerignore complete**    | Exclude node_modules, .git, logs         |
+- **COPY package.json first**: Deps layer cached until deps change
+- **Combine RUN commands**: Fewer layers = smaller image
+- **Clean in same RUN**: `npm ci && npm cache clean` not separate
+- **Order by change frequency**: Least changing → Most changing
+- **.dockerignore complete**: Exclude node_modules, .git, logs
 
 ---
 
 ## Security Hardening Checklist
 
-| Security Rule                | Implementation                      |
-| ---------------------------- | ----------------------------------- |
-| **Non-root user**            | `adduser` + `USER 1001`             |
-| **Pin base image versions**  | `node:20.10.0-alpine`, not `latest` |
-| **No secrets in Dockerfile** | Use build secrets or runtime env    |
-| **Minimal packages**         | No vim, curl unless needed          |
-| **Read-only filesystem**     | `--read-only` runtime flag          |
-| **Scan for vulnerabilities** | `docker scout`, `trivy`             |
+- **Non-root user**: `adduser` + `USER 1001`
+- **Pin base image versions**: `node:20.10.0-alpine`, not `latest`
+- **No secrets in Dockerfile**: Use build secrets or runtime env
+- **Minimal packages**: No vim, curl unless needed
+- **Read-only filesystem**: `--read-only` runtime flag
+- **Scan for vulnerabilities**: `docker scout`, `trivy`
 
 ### Build Secrets Pattern (BuildKit)
 
@@ -185,17 +188,15 @@ Need shell for debugging?
 
 ## Anti-Patterns (DON'T)
 
-| ❌ Anti-Pattern                | ✅ Correct Approach                       |
-| ------------------------------ | ----------------------------------------- |
-| `FROM node:latest`             | `FROM node:20.10.0-alpine`                |
-| `RUN npm install` (not ci)     | `RUN npm ci --only=production`            |
-| Running as root                | Create user, `USER 1001`                  |
-| Secrets in ENV                 | Build secrets or external secrets manager |
-| `COPY . .` before package.json | Copy package.json first for caching       |
-| Separate RUN for cleanup       | Clean in same RUN command                 |
-| No .dockerignore               | Comprehensive .dockerignore               |
-| No health checks               | HEALTHCHECK instruction                   |
-| 1GB+ images                    | Multi-stage + alpine base                 |
+- `FROM node:latest`: `FROM node:20.10.0-alpine`
+- `RUN npm install` (not ci): `RUN npm ci --only=production`
+- Running as root: Create user, `USER 1001`
+- Secrets in ENV: Build secrets or external secrets manager
+- `COPY . .` before package.json: Copy package.json first for caching
+- Separate RUN for cleanup: Clean in same RUN command
+- No .dockerignore: Comprehensive .dockerignore
+- No health checks: HEALTHCHECK instruction
+- 1GB+ images: Multi-stage + alpine base
 
 ---
 
@@ -214,26 +215,22 @@ Need shell for debugging?
 
 ## 🔴 Self-Check Before Completing
 
-| Check                   | Question                                   |
-| ----------------------- | ------------------------------------------ |
-| ✅ **Multi-stage?**     | Is build separated from runtime?           |
-| ✅ **Non-root?**        | Container runs as non-root user?           |
-| ✅ **Pinned versions?** | Base image and deps have explicit version? |
-| ✅ **Health check?**    | HEALTHCHECK or compose healthcheck?        |
-| ✅ **Secrets safe?**    | No secrets in layers or ENV?               |
-| ✅ **Image size?**      | Under 500MB for typical apps?              |
-| ✅ **.dockerignore?**   | Excludes node_modules, .git, logs?         |
+- ✅ **Multi-stage?**: Is build separated from runtime?
+- ✅ **Non-root?**: Container runs as non-root user?
+- ✅ **Pinned versions?**: Base image and deps have explicit version?
+- ✅ **Health check?**: HEALTHCHECK or compose healthcheck?
+- ✅ **Secrets safe?**: No secrets in layers or ENV?
+- ✅ **Image size?**: Under 500MB for typical apps?
+- ✅ **.dockerignore?**: Excludes node_modules, .git, logs?
 
 ---
 
 ## Related Skills
 
-| Need                  | Skill                     |
-| --------------------- | ------------------------- |
-| CI/CD pipelines       | `github-actions` (future) |
-| Kubernetes deployment | `kubernetes-patterns`     |
-| Server management     | `server-management`       |
-| Deployment workflows  | `deployment-procedures`   |
+- CI/CD pipelines: `github-actions` (future)
+- Kubernetes deployment: `kubernetes-patterns`
+- Server management: `server-management`
+- Deployment workflows: `deployment-procedures`
 
 ---
 

package/kits/coder/skills/documentation-templates/SKILL.md

@@ -6,6 +6,18 @@ allowed-tools: Read, Write, Edit, Glob, Grep, Bash
 
 # Documentation Templates
 
+## ⚡ Quick Reference
+
+- **README must-haves**: Title+tagline · Quick Start (< 5min) · Installation · Features · License
+- **Docs-as-Code**: In same repo as code · PR-reviewed · Auto-deployed · Versioned with releases
+- **API docs**: Every endpoint: method+path+params+request+response+errors+examples
+- **ADR format**: Title · Status · Context · Decision · Consequences · Date
+- **Code comments**: WHY not WHAT · Complex algorithms only · API contracts · Gotchas
+- **AI-friendly**: Structured H1-H3 · Self-contained sections · JSON examples · Mermaid diagrams
+
+---
+
+
 > Good documentation is the difference between a project that gets adopted and one that gets abandoned.
 
 ---
@@ -417,24 +429,20 @@ docs/
 
 ## ❌ Anti-Patterns
 
-| ❌ Don't                        | ✅ Do                           |
-| ------------------------------- | ------------------------------- |
-| Document implementation details | Document behavior and contracts |
-| Write docs after release        | Write docs during development   |
-| Skip error documentation        | Document all error cases        |
-| Use outdated examples           | Test examples in CI             |
-| Wall of text without structure  | Use headers, lists, code blocks |
-| Assume reader knows context     | Explain the "why" first         |
+- Document implementation details: Document behavior and contracts
+- Write docs after release: Write docs during development
+- Skip error documentation: Document all error cases
+- Use outdated examples: Test examples in CI
+- Wall of text without structure: Use headers, lists, code blocks
+- Assume reader knows context: Explain the "why" first
 
 ---
 
 ## 🔗 Related Skills
 
-| Need         | Skill              |
-| ------------ | ------------------ |
-| API design   | `api-patterns`     |
-| Clean code   | `clean-code`       |
-| Architecture | `mermaid-diagrams` |
+- API design: `api-patterns`
+- Clean code: `clean-code`
+- Architecture: `mermaid-diagrams`
 
 ---
 

package/kits/coder/skills/e2e-testing/SKILL.md

@@ -6,17 +6,27 @@ allowed-tools: Read, Write, Edit, Glob, Grep, Bash
 
 # E2E Testing Patterns
 
+## ⚡ Quick Reference
+
+- **Tool selection**: Playwright (recommended) for modern · Cypress for existing projects · Avoid Selenium
+- **Code critical paths only**: Login · Checkout · Core flows · Not every UI state
+- **Selectors**: `data-testid` attributes · `getByRole` · Never CSS/XPath class selectors (brittle)
+- **Reliability**: `await expect(locator).toBeVisible()` not `sleep()` · Retry-aware assertions · Isolated test data
+- **CI**: Parallel sharding · Retry flaky tests once · Store traces/screenshots on failure
+- **Page Objects**: Group locators and actions per page · Reusable · Separate from test logic
+
+---
+
+
 > **Philosophy:** E2E tests should give confidence to ship, not slow you down. Test critical paths, make them reliable, run them fast.
 
 ---
 
 ## 📑 Content Map
 
-| File                             | When to Read                 |
-| -------------------------------- | ---------------------------- |
-| `references/playwright-api.md`   | Playwright-specific patterns |
-| `references/cypress-patterns.md` | Cypress-specific patterns    |
-| `references/ci-integration.md`   | CI/CD setup and optimization |
+- `references/playwright-api.md`: Playwright-specific patterns
+- `references/cypress-patterns.md`: Cypress-specific patterns
+- `references/ci-integration.md`: CI/CD setup and optimization
 
 ---
 
@@ -24,23 +34,19 @@ allowed-tools: Read, Write, Edit, Glob, Grep, Bash
 
 ### ✅ Use E2E When
 
-| Scenario                    | Why E2E                     |
-| --------------------------- | --------------------------- |
-| Critical user journeys      | Full flow validation        |
-| Authentication flows        | Security-critical paths     |
-| Payment/checkout            | Money-critical paths        |
-| Cross-browser compatibility | Real browser behavior       |
-| Accessibility validation    | Full page context needed    |
-| Visual regression           | Layout/styling verification |
+- Critical user journeys: Full flow validation
+- Authentication flows: Security-critical paths
+- Payment/checkout: Money-critical paths
+- Cross-browser compatibility: Real browser behavior
+- Accessibility validation: Full page context needed
+- Visual regression: Layout/styling verification
 
 ### ❌ Don't Use E2E When
 
-| Scenario                        | Better Alternative                  |
-| ------------------------------- | ----------------------------------- |
-| Unit logic testing              | Unit tests (Vitest, Jest)           |
-| Component behavior              | Integration tests (Testing Library) |
-| API contract validation         | API tests (Supertest, Hurl)         |
-| Unstable/frequently changing UI | Lower-level tests first             |
+- Unit logic testing: Unit tests (Vitest, Jest)
+- Component behavior: Integration tests (Testing Library)
+- API contract validation: API tests (Supertest, Hurl)
+- Unstable/frequently changing UI: Lower-level tests first
 
 ---
 
@@ -98,11 +104,9 @@ test("successful login redirects to dashboard", async ({ page }) => {
 
 ### Test Structure (AAA Pattern)
 
-| Phase       | Action                     |
-| ----------- | -------------------------- |
-| **Arrange** | Navigate, set up test data |
-| **Act**     | Perform user actions       |
-| **Assert**  | Verify expected outcomes   |
+- **Arrange**: Navigate, set up test data
+- **Act**: Perform user actions
+- **Assert**: Verify expected outcomes
 
 ---
 
@@ -120,12 +124,10 @@ test("successful login redirects to dashboard", async ({ page }) => {
 
 ### ❌ Avoid
 
-| Selector      | Why                         |
-| ------------- | --------------------------- |
-| CSS class     | Styling changes break tests |
-| Tag name only | Too generic, fragile        |
-| XPath         | Hard to read, fragile       |
-| nth-child     | Order-dependent, fragile    |
+- CSS class: Styling changes break tests
+- Tag name only: Too generic, fragile
+- XPath: Hard to read, fragile
+- nth-child: Order-dependent, fragile
 
 ---
 
@@ -144,12 +146,10 @@ await expect(page.getByTestId("data")).toContainText("Loaded");
 
 ### Handling Flaky Tests
 
-| Symptom                | Solution                                |
-| ---------------------- | --------------------------------------- |
-| Element not found      | Use auto-wait, increase timeout         |
-| Race conditions        | Wait for network idle, specific element |
-| Animation interference | Disable animations in CI                |
-| Timing issues          | Use explicit waits, not setTimeout      |
+- Element not found: Use auto-wait, increase timeout
+- Race conditions: Wait for network idle, specific element
+- Animation interference: Disable animations in CI
+- Timing issues: Use explicit waits, not setTimeout
 
 ```typescript
 // ✅ Wait for network idle
@@ -337,13 +337,11 @@ jobs:
 
 ### CI Optimization
 
-| Technique        | Impact                           |
-| ---------------- | -------------------------------- |
-| Parallel workers | 2-4x faster                      |
-| Sharding         | Distribute across machines       |
-| Caching browsers | Faster setup                     |
-| Headless mode    | Required in CI                   |
-| Fail fast        | Stop on first failure (optional) |
+- Parallel workers: 2-4x faster
+- Sharding: Distribute across machines
+- Caching browsers: Faster setup
+- Headless mode: Required in CI
+- Fail fast: Stop on first failure (optional)
 
 ```typescript
 // playwright.config.ts
@@ -402,15 +400,13 @@ test("debug example", async ({ page }) => {
 
 ## 11. Anti-Patterns
 
-| ❌ Don't                    | ✅ Do                             |
-| --------------------------- | --------------------------------- |
-| Test implementation details | Test user-visible behavior        |
-| Use fixed `setTimeout`      | Use explicit waits                |
-| Share state between tests   | Isolate each test                 |
-| Test everything in E2E      | Use testing pyramid               |
-| Skip flaky tests            | Fix root cause                    |
-| Hard-coded credentials      | Environment variables, test users |
-| Run against production      | Use staging/preview environments  |
+- Test implementation details: Test user-visible behavior
+- Use fixed `setTimeout`: Use explicit waits
+- Share state between tests: Isolate each test
+- Test everything in E2E: Use testing pyramid
+- Skip flaky tests: Fix root cause
+- Hard-coded credentials: Environment variables, test users
+- Run against production: Use staging/preview environments
 
 ---
 
@@ -445,12 +441,10 @@ test("user registration", async ({ page }) => {
 
 ## Related Skills
 
-| Need                     | Skill                   |
-| ------------------------ | ----------------------- |
-| Unit/integration testing | `testing-patterns`      |
-| CI/CD pipelines          | `github-actions`        |
-| Performance testing      | `performance-profiling` |
-| Accessibility            | `web-design-guidelines` |
+- Unit/integration testing: `testing-patterns`
+- CI/CD pipelines: `github-actions`
+- Performance testing: `performance-profiling`
+- Accessibility: `web-design-guidelines`
 
 ---
 

package/kits/coder/skills/flutter-patterns/SKILL.md

@@ -8,19 +8,27 @@ priority: MEDIUM
 
 # Flutter Patterns - Cross-Platform Development Excellence
 
-> **Philosophy:** Build once, run anywhere with native performance. Widget composition over inheritance, const constructors everywhere.
+## ⚡ Quick Reference
+
+- **Widgets**: `const` constructors everywhere · `StatelessWidget` by default · `StatefulWidget` only for local state
+- **State**: Local → `setState` · Feature-wide → Riverpod · App-wide → Riverpod Provider
+- **Performance**: `const` widgets skip rebuilds · `RepaintBoundary` for heavy paths · ListView.builder for long lists
+- **Navigation**: GoRouter for complex routing · Named routes · Pass data via constructor not global
+- **Async**: `FutureBuilder`/`StreamBuilder` · Handle loading/error states always · `ref.watch()` for Riverpod
+- **Testing**: Widget tests for UI · Unit tests for logic · Integration tests for flows on real device
+
+---
+
 
 ---
 
 ## 🎯 Core Principles
 
-| Principle          | Rule                                             |
-| ------------------ | ------------------------------------------------ |
-| **Composition**    | Compose widgets, don't inherit                   |
-| **Immutability**   | Use `const` constructors for optimal performance |
-| **Null Safety**    | Dart 3 null safety is mandatory                  |
-| **Platform Aware** | One codebase, platform-specific polish           |
-| **Test First**     | Widget tests, integration tests, golden tests    |
+- **Composition**: Compose widgets, don't inherit
+- **Immutability**: Use `const` constructors for optimal performance
+- **Null Safety**: Dart 3 null safety is mandatory
+- **Platform Aware**: One codebase, platform-specific polish
+- **Test First**: Widget tests, integration tests, golden tests
 
 ```
 ❌ WRONG: Create God widgets with everything
@@ -152,12 +160,10 @@ class ActionButton extends StatelessWidget {
 
 ### Widget Keys
 
-| Scenario                   | Key Type          |
-| -------------------------- | ----------------- |
-| List items with unique IDs | `ValueKey(id)`    |
-| Animated list items        | `ObjectKey(item)` |
-| Form fields                | `GlobalKey`       |
-| Never needs identity       | No key            |
+- List items with unique IDs: `ValueKey(id)`
+- Animated list items: `ObjectKey(item)`
+- Form fields: `GlobalKey`
+- Never needs identity: No key
 
 ```dart
 ListView.builder(
@@ -177,13 +183,11 @@ ListView.builder(
 
 ### Widget Rebuild Prevention
 
-| Technique            | When to Use                        |
-| -------------------- | ---------------------------------- |
-| `const` constructors | Always when possible               |
-| `Consumer` widgets   | Scope Riverpod rebuilds            |
-| `select` in Riverpod | Watch only needed parts            |
-| `RepaintBoundary`    | Isolate expensive paint operations |
-| `ListView.builder`   | Large lists (virtualized)          |
+- `const` constructors: Always when possible
+- `Consumer` widgets: Scope Riverpod rebuilds
+- `select` in Riverpod: Watch only needed parts
+- `RepaintBoundary`: Isolate expensive paint operations
+- `ListView.builder`: Large lists (virtualized)
 
 ### Image Optimization
 
@@ -289,41 +293,35 @@ if (kIsWeb) {
 
 ## 🚨 Anti-Patterns
 
-| ❌ Don't                         | ✅ Do                                 |
-| -------------------------------- | ------------------------------------- |
-| Inherit from StatelessWidget/ful | Compose widgets                       |
-| Skip `const` constructors        | Add `const` everywhere possible       |
-| setState in large widgets        | Use state management solution         |
-| Fetch data in build()            | Use FutureBuilder or state management |
-| Large build() methods            | Extract smaller widget functions      |
-| Hardcode colors/sizes            | Use Theme and design tokens           |
-| Skip null safety                 | Embrace Dart 3 null safety            |
-| Test only happy paths            | Test edge cases and errors            |
+- Inherit from StatelessWidget/ful: Compose widgets
+- Skip `const` constructors: Add `const` everywhere possible
+- setState in large widgets: Use state management solution
+- Fetch data in build(): Use FutureBuilder or state management
+- Large build() methods: Extract smaller widget functions
+- Hardcode colors/sizes: Use Theme and design tokens
+- Skip null safety: Embrace Dart 3 null safety
+- Test only happy paths: Test edge cases and errors
 
 ---
 
 ## ✅ Self-Check Before Completing
 
-| Check               | Question                                |
-| ------------------- | --------------------------------------- |
-| ✅ **Const?**       | Are const constructors used everywhere? |
-| ✅ **Keys?**        | Do list items have stable keys?         |
-| ✅ **State?**       | Is state management properly scoped?    |
-| ✅ **Null safe?**   | Is null safety properly handled?        |
-| ✅ **Tested?**      | Are widgets and logic tested?           |
-| ✅ **Platform?**    | Tested on both iOS and Android?         |
-| ✅ **Performance?** | No jank, smooth scrolling?              |
+- ✅ **Const?**: Are const constructors used everywhere?
+- ✅ **Keys?**: Do list items have stable keys?
+- ✅ **State?**: Is state management properly scoped?
+- ✅ **Null safe?**: Is null safety properly handled?
+- ✅ **Tested?**: Are widgets and logic tested?
+- ✅ **Platform?**: Tested on both iOS and Android?
+- ✅ **Performance?**: No jank, smooth scrolling?
 
 ---
 
 ## 🔗 Related Skills
 
-| Need                    | Skill                   |
-| ----------------------- | ----------------------- |
-| React Native comparison | `react-native-patterns` |
-| Mobile design           | `mobile-design`         |
-| Testing patterns        | `testing-patterns`      |
-| Performance profiling   | `performance-profiling` |
+- React Native comparison: `react-native-patterns`
+- Mobile design: `mobile-design`
+- Testing patterns: `testing-patterns`
+- Performance profiling: `performance-profiling`
 
 ---
 

package/kits/coder/skills/frontend-design/SKILL.md

@@ -6,6 +6,18 @@ allowed-tools: Read, Write, Edit, Glob, Grep, Bash
 
 # Frontend Design Patterns
 
+## ⚡ Quick Reference
+
+- **Color**: WCAG 4.5:1 contrast · HSL for tints/shades · 60-30-10 rule (primary/secondary/accent)
+- **Typography**: 16px base · `rem` units · 1.5 line-height body · max 75 chars per line
+- **Spacing**: 4px base unit (4/8/16/24/32/48/64/96) · consistent scale · CSS custom properties
+- **Layout**: Mobile-first · Grid for 2D · Flexbox for 1D · `max-width: 1200px` containers
+- **Micro-interactions**: Hover 150ms ease · Focus rings visible · Loading skeleton not spinner
+- **Dark mode**: CSS custom properties swap · `prefers-color-scheme` media query as default
+
+---
+
+
 > Design is how it works.
 
 ---
@@ -23,14 +35,12 @@ allowed-tools: Read, Write, Edit, Glob, Grep, Bash
 
 ### Palette Structure
 
-| Role      | Use Case                    |
-| --------- | --------------------------- |
-| Primary   | CTAs, links, brand elements |
-| Secondary | Supporting elements         |
-| Neutral   | Text, borders, backgrounds  |
-| Success   | Confirmations, completed    |
-| Warning   | Caution, pending            |
-| Error     | Errors, destructive actions |
+- Primary: CTAs, links, brand elements
+- Secondary: Supporting elements
+- Neutral: Text, borders, backgrounds
+- Success: Confirmations, completed
+- Warning: Caution, pending
+- Error: Errors, destructive actions
 
 ### Contrast (WCAG)
 
@@ -78,21 +88,17 @@ allowed-tools: Read, Write, Edit, Glob, Grep, Bash
 
 ## 📊 Layout Patterns
 
-| Pattern    | Description                   |
-| ---------- | ----------------------------- |
-| Holy Grail | Header, footer, 3-col middle  |
-| Sidebar    | Fixed sidebar + fluid content |
-| Dashboard  | Sidebar nav + header + grid   |
-| Cards      | Auto-filling responsive grid  |
+- Holy Grail: Header, footer, 3-col middle
+- Sidebar: Fixed sidebar + fluid content
+- Dashboard: Sidebar nav + header + grid
+- Cards: Auto-filling responsive grid
 
 ---
 
 ## ✨ Micro-Interactions
 
-| Principle | Implementation                   |
-| --------- | -------------------------------- |
-| Duration  | 150-300ms for interactions       |
-| Easing    | ease-out for enter, ease-in exit |
+- Duration: 150-300ms for interactions
+- Easing: ease-out for enter, ease-in exit
 
 ---
 
@@ -109,12 +115,10 @@ allowed-tools: Read, Write, Edit, Glob, Grep, Bash
 
 ## ❌ Anti-Patterns
 
-| Don't                    | Do                   |
-| ------------------------ | -------------------- |
-| Low contrast text        | WCAG AA minimum      |
-| More than 3 fonts        | 1-2 fonts            |
-| Inconsistent spacing     | Use spacing tokens   |
-| Animation for decoration | Animate purposefully |
+- Low contrast text: WCAG AA minimum
+- More than 3 fonts: 1-2 fonts
+- Inconsistent spacing: Use spacing tokens
+- Animation for decoration: Animate purposefully
 
 ---