@harness-engineering/cli 1.23.2 → 1.24.0

package/dist/agents/skills/codex/a11y-modal-patterns/SKILL.md

@@ -0,0 +1,201 @@
+# Accessible Modal Dialog Patterns
+
+> Build accessible modal dialogs with focus trapping, escape dismissal, background inertness, and screen reader announcements
+
+## When to Use
+
+- Implementing modal dialogs, confirmation prompts, or alert dialogs
+- Building sheet/drawer overlays that require user attention
+- Creating image lightboxes or video players in overlay
+- Reviewing existing modals for keyboard and screen reader accessibility
+
+## Instructions
+
+1. **Use the native `<dialog>` element when possible.** Modern browsers support `<dialog>` with `.showModal()`, which provides built-in focus trapping, Escape dismissal, and backdrop styling.
+
+```tsx
+function Modal({ isOpen, onClose, children }: ModalProps) {
+  const dialogRef = useRef<HTMLDialogElement>(null);
+
+  useEffect(() => {
+    const dialog = dialogRef.current;
+    if (!dialog) return;
+    if (isOpen) {
+      dialog.showModal();
+    } else {
+      dialog.close();
+    }
+  }, [isOpen]);
+
+  return (
+    <dialog ref={dialogRef} onClose={onClose} aria-labelledby="dialog-title">
+      {children}
+    </dialog>
+  );
+}
+```
+
+2. **If using a custom modal (not `<dialog>`), implement all accessibility requirements manually:**
+
+```tsx
+function CustomModal({ isOpen, onClose, title, children }: ModalProps) {
+  const modalRef = useRef<HTMLDivElement>(null);
+  const triggerRef = useRef<HTMLElement | null>(null);
+
+  useEffect(() => {
+    if (isOpen) {
+      triggerRef.current = document.activeElement as HTMLElement;
+      // Focus the first focusable element in the modal
+      requestAnimationFrame(() => {
+        const firstFocusable = modalRef.current?.querySelector<HTMLElement>(
+          'button, [href], input, select, textarea, [tabindex]:not([tabindex="-1"])'
+        );
+        firstFocusable?.focus();
+      });
+    }
+    return () => {
+      // Return focus to trigger when closing
+      triggerRef.current?.focus();
+    };
+  }, [isOpen]);
+
+  if (!isOpen) return null;
+
+  return (
+    <>
+      <div className="backdrop" onClick={onClose} />
+      <div
+        ref={modalRef}
+        role="dialog"
+        aria-modal="true"
+        aria-labelledby="dialog-title"
+        onKeyDown={(e) => {
+          if (e.key === 'Escape') onClose();
+        }}
+      >
+        <h2 id="dialog-title">{title}</h2>
+        {children}
+      </div>
+    </>
+  );
+}
+```
+
+3. **Trap focus inside the modal.** Tab and Shift+Tab must cycle through focusable elements within the modal only. Use a focus trap library or implement manually.
+
+```typescript
+function useFocusTrap(ref: RefObject<HTMLElement>, active: boolean) {
+  useEffect(() => {
+    if (!active || !ref.current) return;
+    const container = ref.current;
+
+    const handleKeyDown = (e: KeyboardEvent) => {
+      if (e.key !== 'Tab') return;
+
+      const focusable = container.querySelectorAll<HTMLElement>(
+        'a[href], button:not([disabled]), input:not([disabled]), select, textarea, [tabindex]:not([tabindex="-1"])'
+      );
+      const first = focusable[0];
+      const last = focusable[focusable.length - 1];
+
+      if (e.shiftKey && document.activeElement === first) {
+        e.preventDefault();
+        last.focus();
+      } else if (!e.shiftKey && document.activeElement === last) {
+        e.preventDefault();
+        first.focus();
+      }
+    };
+
+    container.addEventListener('keydown', handleKeyDown);
+    return () => container.removeEventListener('keydown', handleKeyDown);
+  }, [ref, active]);
+}
+```
+
+4. **Make background content inert.** When the modal is open, background content must not be interactive. Use the `inert` attribute on the main content.
+
+```tsx
+useEffect(() => {
+  const main = document.getElementById('app-root');
+  if (isOpen && main) {
+    main.setAttribute('inert', '');
+    return () => main.removeAttribute('inert');
+  }
+}, [isOpen]);
+```
+
+5. **Close on Escape key press.** This is expected behavior — users should never be trapped in a dialog without a keyboard exit.
+
+6. **Close on backdrop click** for non-critical dialogs. For confirmation dialogs (`role="alertdialog"`), require explicit button interaction.
+
+7. **Use `role="alertdialog"` for confirmation prompts** that require the user to acknowledge or make a choice. These should not close on backdrop click.
+
+```tsx
+<div
+  role="alertdialog"
+  aria-modal="true"
+  aria-labelledby="alert-title"
+  aria-describedby="alert-desc"
+>
+  <h2 id="alert-title">Delete Account?</h2>
+  <p id="alert-desc">
+    This will permanently delete your account and all data. This cannot be undone.
+  </p>
+  <button onClick={onCancel}>Cancel</button>
+  <button onClick={onConfirm}>Delete Account</button>
+</div>
+```
+
+8. **Return focus to the trigger element when the modal closes.** Save a reference to `document.activeElement` before opening the modal and call `.focus()` on it after closing.
+
+9. **Prevent body scroll when modal is open.**
+
+```css
+body.modal-open {
+  overflow: hidden;
+}
+```
+
+## Details
+
+**`<dialog>` vs. custom modal:** The native `<dialog>` element with `.showModal()` provides focus trapping, Escape dismissal, `::backdrop` styling, top-layer rendering, and `inert` behavior on background content — all for free. It is supported in all modern browsers. Use custom implementations only when you need behavior that `<dialog>` does not support.
+
+**Focus management sequence:**
+
+1. User clicks trigger button
+2. Save reference to trigger element
+3. Open modal, move focus to first focusable element (or the modal container with `tabIndex={-1}`)
+4. Trap focus within modal
+5. User closes modal (Escape, close button, or backdrop click)
+6. Return focus to saved trigger element
+
+**`aria-modal="true"` vs. `inert`:** `aria-modal="true"` tells screen readers that content behind the dialog is not interactive. However, some screen readers do not fully respect this. Adding `inert` to background content provides a robust fallback that works at the browser level.
+
+**Common mistakes:**
+
+- Focus escaping the modal when Tab reaches the end (no focus trap)
+- Not returning focus to trigger on close (user is lost on the page)
+- Background content scrollable and interactive while modal is open
+- Missing accessible name (`aria-labelledby` or `aria-label`)
+- Auto-focusing a close button instead of the most relevant element
+
+## Source
+
+https://www.w3.org/WAI/ARIA/apd/patterns/dialog-modal/
+
+## Process
+
+1. Read the instructions and examples in this document.
+2. Apply the patterns to your implementation, adapting to your specific context.
+3. Verify your implementation against the details and edge cases listed above.
+
+## Harness Integration
+
+- **Type:** knowledge — this skill is a reference document, not a procedural workflow.
+- **No tools or state** — consumed as context by other skills and agents.
+
+## Success Criteria
+
+- The patterns described in this document are applied correctly in the implementation.
+- Edge cases and anti-patterns listed in this document are avoided.

package/dist/agents/skills/codex/a11y-modal-patterns/skill.yaml

@@ -0,0 +1,42 @@
+name: a11y-modal-patterns
+version: '1.0.0'
+description: Build accessible modal dialogs with focus trapping, escape dismissal, and screen reader announcements
+cognitive_mode: advisory-guide
+type: knowledge
+tier: 3
+triggers:
+  - manual
+platforms:
+  - claude-code
+  - gemini-cli
+  - cursor
+  - codex
+tools: []
+paths:
+  - '**/*.tsx'
+  - '**/*.html'
+  - '**/*.vue'
+  - '**/*.svelte'
+  - '**/*.astro'
+related_skills:
+  - a11y-keyboard-navigation
+  - a11y-aria-patterns
+  - a11y-semantic-html
+stack_signals:
+  - accessibility
+  - react
+  - typescript
+keywords:
+  - modal
+  - dialog
+  - focus-trap
+  - inert
+  - overlay
+  - dismiss
+metadata:
+  author: community
+  upstream: w3.org/WAI/ARIA/apd/patterns/dialog-modal
+state:
+  persistent: false
+  files: []
+depends_on: []

package/dist/agents/skills/codex/a11y-motion-animation/SKILL.md

@@ -0,0 +1,199 @@
+# Motion and Animation Accessibility
+
+> Implement animations that respect user motion preferences, avoid seizure triggers, and provide pause controls
+
+## When to Use
+
+- Adding animations, transitions, or motion effects to a UI
+- Building auto-playing carousels, videos, or background animations
+- Implementing page transitions in a SPA
+- Using parallax scrolling or scroll-triggered animations
+- Reviewing existing animations for accessibility compliance
+
+## Instructions
+
+1. **Respect `prefers-reduced-motion`.** Users who enable "Reduce motion" in their OS settings have vestibular disorders or motion sensitivity. Query this preference and eliminate or simplify animations.
+
+```css
+/* Default: full animations */
+.card {
+  transition:
+    transform 0.3s ease,
+    opacity 0.3s ease;
+}
+
+.card:hover {
+  transform: scale(1.05);
+}
+
+/* Reduced motion: remove or simplify */
+@media (prefers-reduced-motion: reduce) {
+  .card {
+    transition: none;
+  }
+  .card:hover {
+    transform: none;
+  }
+}
+```
+
+2. **Use a motion-first or reduce-first strategy.** Choose one approach and apply it consistently:
+
+```css
+/* Approach A: motion-first (add animations, remove for reduced-motion) */
+.element {
+  animation: slide-in 0.5s ease;
+}
+@media (prefers-reduced-motion: reduce) {
+  .element {
+    animation: none;
+  }
+}
+
+/* Approach B: reduce-first (no animations by default, add for no-preference) */
+.element {
+  animation: none;
+}
+@media (prefers-reduced-motion: no-preference) {
+  .element {
+    animation: slide-in 0.5s ease;
+  }
+}
+```
+
+Approach B is safer — users get no animation by default.
+
+3. **In JavaScript, check the preference before triggering animations.**
+
+```typescript
+const prefersReducedMotion = window.matchMedia('(prefers-reduced-motion: reduce)').matches;
+
+function animateElement(element: HTMLElement) {
+  if (prefersReducedMotion) {
+    // Instant state change, no animation
+    element.style.opacity = '1';
+    return;
+  }
+  element.animate([{ opacity: 0 }, { opacity: 1 }], { duration: 300, easing: 'ease-in' });
+}
+```
+
+4. **In React, create a hook for motion preferences.**
+
+```typescript
+function usePrefersReducedMotion(): boolean {
+  const [prefersReduced, setPrefersReduced] = useState(
+    () => window.matchMedia('(prefers-reduced-motion: reduce)').matches
+  );
+
+  useEffect(() => {
+    const mql = window.matchMedia('(prefers-reduced-motion: reduce)');
+    const handler = (e: MediaQueryListEvent) => setPrefersReduced(e.matches);
+    mql.addEventListener('change', handler);
+    return () => mql.removeEventListener('change', handler);
+  }, []);
+
+  return prefersReduced;
+}
+
+// Usage
+function AnimatedCard() {
+  const reduceMotion = usePrefersReducedMotion();
+  return (
+    <motion.div
+      initial={{ opacity: 0, y: reduceMotion ? 0 : 20 }}
+      animate={{ opacity: 1, y: 0 }}
+      transition={{ duration: reduceMotion ? 0 : 0.3 }}
+    />
+  );
+}
+```
+
+5. **Provide pause, stop, or hide controls for auto-playing content.** Any content that moves, blinks, or auto-updates for more than 5 seconds must have user controls (WCAG 2.2.2).
+
+```tsx
+function AutoCarousel({ slides }: { slides: Slide[] }) {
+  const [isPaused, setIsPaused] = useState(false);
+
+  return (
+    <div>
+      <button
+        onClick={() => setIsPaused(!isPaused)}
+        aria-label={isPaused ? 'Play slideshow' : 'Pause slideshow'}
+      >
+        {isPaused ? <PlayIcon /> : <PauseIcon />}
+      </button>
+      <Carousel autoPlay={!isPaused} slides={slides} />
+    </div>
+  );
+}
+```
+
+6. **Avoid flashing content.** Content that flashes more than 3 times per second can trigger seizures (WCAG 2.3.1 — Three Flashes). This is a hard constraint with no workaround.
+
+7. **Avoid large-scale motion.** Parallax effects, zooming transitions, and full-page scroll animations are common vestibular triggers. When using these, always check `prefers-reduced-motion` and provide a static alternative.
+
+8. **Keep essential animations short.** Transitions that communicate state changes (loading, success, error) should be under 500ms. Long, elaborate animations add cognitive load and frustrate users who interact frequently.
+
+9. **Use `will-change` and GPU-composited properties for performance.** Janky animations that drop frames are worse than no animation. Stick to `transform` and `opacity` for smooth 60fps animations.
+
+```css
+.animated {
+  will-change: transform, opacity;
+  transition:
+    transform 0.2s ease,
+    opacity 0.2s ease;
+}
+```
+
+## Details
+
+**WCAG requirements:**
+
+- **2.2.2 Pause, Stop, Hide:** Auto-playing content that lasts more than 5 seconds must have controls
+- **2.3.1 Three Flashes:** No content flashes more than 3 times per second
+- **2.3.3 Animation from Interactions (AAA):** Motion triggered by user interaction can be disabled
+
+**What counts as "reduced motion":**
+
+- Remove: parallax scrolling, auto-playing animations, bouncing/shaking effects, zoom transitions
+- Keep: opacity fades (instant or very fast), color changes, layout reflows
+- Simplify: Replace slide transitions with fade, replace bounce with ease
+
+**Framer Motion integration:**
+
+```tsx
+<motion.div
+  layout
+  transition={{
+    layout: { duration: prefersReduced ? 0 : 0.3 },
+  }}
+/>
+```
+
+**Common mistakes:**
+
+- Checking `prefers-reduced-motion` once at load (user can change it at runtime — use `addEventListener`)
+- Removing only CSS animations while JavaScript-driven animations still run
+- Interpreting "reduced motion" as "no animation" (subtle fades and opacity changes are usually fine)
+- Auto-playing video backgrounds without a pause button
+
+## Source
+
+https://www.w3.org/WAI/WCAG21/Understanding/animation-from-interactions
+
+## Process
+
+1. Read the instructions and examples in this document.
+2. Apply the patterns to your implementation, adapting to your specific context.
+3. Verify your implementation against the details and edge cases listed above.
+
+## Harness Integration
+
+- **Type:** knowledge — this skill is a reference document, not a procedural workflow.
+- **No tools or state** — consumed as context by other skills and agents.
+
+## Success Criteria
+
+- The patterns described in this document are applied correctly in the implementation.
+- Edge cases and anti-patterns listed in this document are avoided.

package/dist/agents/skills/codex/a11y-motion-animation/skill.yaml

@@ -0,0 +1,45 @@
+name: a11y-motion-animation
+version: '1.0.0'
+description: Implement animations that respect user motion preferences and avoid triggering vestibular disorders
+cognitive_mode: advisory-guide
+type: knowledge
+tier: 3
+triggers:
+  - manual
+platforms:
+  - claude-code
+  - gemini-cli
+  - cursor
+  - codex
+tools: []
+paths:
+  - '**/*.tsx'
+  - '**/*.html'
+  - '**/*.vue'
+  - '**/*.svelte'
+  - '**/*.astro'
+  - '**/*.css'
+  - '**/*.scss'
+related_skills:
+  - a11y-color-contrast
+  - a11y-semantic-html
+  - design-motion-principles
+  - design-micro-interactions
+stack_signals:
+  - accessibility
+  - react
+  - typescript
+keywords:
+  - motion
+  - animation
+  - reduced-motion
+  - prefers-reduced-motion
+  - vestibular
+  - transitions
+metadata:
+  author: community
+  upstream: w3.org/WAI/WCAG21/Understanding/animation-from-interactions
+state:
+  persistent: false
+  files: []
+depends_on: []

package/dist/agents/skills/codex/a11y-screen-reader-testing/SKILL.md

@@ -0,0 +1,138 @@
+# Screen Reader Testing
+
+> Test web applications with screen readers to verify accessible navigation, announcements, and interaction patterns
+
+## When to Use
+
+- Verifying that ARIA attributes produce the expected screen reader output
+- Testing custom widgets (tabs, comboboxes, dialogs) for assistive technology support
+- Validating focus management in dynamic content (SPA route changes, modals)
+- QA testing before release for accessibility compliance
+- Learning how screen reader users experience your application
+
+## Instructions
+
+1. **Choose the right screen reader for your target platform.** Test with the browser/screen reader combination your users actually use.
+
+| Screen Reader | Platform  | Browser         | Market Share |
+| ------------- | --------- | --------------- | ------------ |
+| NVDA          | Windows   | Firefox, Chrome | ~40%         |
+| JAWS          | Windows   | Chrome, Edge    | ~30%         |
+| VoiceOver     | macOS/iOS | Safari          | ~25%         |
+| TalkBack      | Android   | Chrome          | ~5%          |
+| Narrator      | Windows   | Edge            | ~3%          |
+
+At minimum, test with NVDA + Firefox and VoiceOver + Safari to cover the majority of screen reader users.
+
+2. **Learn essential VoiceOver commands (macOS).**
+
+```
+Enable/Disable:     Cmd + F5
+VoiceOver key (VO): Ctrl + Option
+
+Navigate next:      VO + Right Arrow
+Navigate previous:  VO + Left Arrow
+Activate element:   VO + Space
+Read all:           VO + A
+Open rotor:         VO + U (navigate landmarks, headings, links, forms)
+Next heading:       VO + Cmd + H
+Next landmark:      VO + Cmd + (no direct shortcut — use rotor)
+```
+
+3. **Learn essential NVDA commands (Windows).**
+
+```
+Enable:             Ctrl + Alt + N
+NVDA key:           Insert (or Caps Lock if configured)
+
+Navigate next:      Down Arrow (in browse mode)
+Navigate previous:  Up Arrow
+Activate element:   Enter
+Read all:           NVDA + Down Arrow
+Elements list:      NVDA + F7 (headings, links, landmarks, form fields)
+Next heading:       H
+Next landmark:      D
+Next form field:    F
+Toggle browse/focus: NVDA + Space
+```
+
+4. **Follow a testing checklist for each page:**
+
+**Landmarks and structure:**
+
+- [ ] Page has a descriptive `<title>`
+- [ ] Landmarks are announced (navigation, main, banner, contentinfo)
+- [ ] Heading hierarchy is logical (h1 > h2 > h3, no skipped levels)
+- [ ] Headings accurately describe their sections
+
+**Navigation:**
+
+- [ ] Skip link works and moves focus to main content
+- [ ] Tab order matches visual order
+- [ ] All interactive elements are reachable by keyboard
+- [ ] Focus indicator is visible on every focusable element
+
+**Forms:**
+
+- [ ] Every input has an announced label
+- [ ] Required fields are announced as required
+- [ ] Error messages are announced when they appear
+- [ ] Form groups are announced with their legend
+
+**Dynamic content:**
+
+- [ ] Modal dialogs announce their title on open
+- [ ] Focus is trapped inside modals
+- [ ] Focus returns to trigger on modal close
+- [ ] Live region updates are announced (toasts, status changes)
+- [ ] SPA route changes announce the new page title or heading
+
+**Images and media:**
+
+- [ ] Informative images have descriptive alt text
+- [ ] Decorative images are skipped
+- [ ] Videos have captions and audio description
+
+5. **Test common user flows, not just individual elements.** Complete a full task: sign up, make a purchase, change settings. Screen reader issues often emerge in transitions between pages and states.
+
+6. **Listen, do not just read the accessibility tree.** Actually listen to the screen reader output. The accessibility tree shows what is exposed, but the spoken output reveals timing, ordering, and verbosity issues that the tree does not show.
+
+7. **Test with the screen reader's virtual/browse mode and forms/focus mode.** NVDA and JAWS have two modes — browse mode (arrow keys navigate the page) and focus mode (arrow keys interact with form controls). Custom widgets must work in both modes or correctly trigger mode switching.
+
+8. **Document screen reader bugs with reproduction steps.** Include: screen reader + version, browser + version, OS, exact steps, expected announcement, actual announcement.
+
+## Details
+
+**VoiceOver Rotor:** Press VO + U to open the rotor — a navigation menu that lists all headings, links, landmarks, and form controls on the page. This is how screen reader users get an overview of page structure. If your headings and landmarks are poor, the rotor is useless.
+
+**Browse mode vs. focus mode:** In browse mode, the screen reader intercepts keystrokes (H jumps to next heading, F jumps to next form field). In focus mode, keystrokes go to the web page (arrow keys navigate within a dropdown). Interactive widgets (comboboxes, grids) need `role` attributes that trigger automatic mode switching.
+
+**Common issues found during testing:**
+
+- Buttons announced as "clickable" instead of "button" (missing semantic element)
+- Images announced by filename (missing `alt`)
+- Form inputs announced without labels (missing `htmlFor` association)
+- Modal content announced mixed with background content (missing `aria-modal` or `inert`)
+- Live region changes not announced (region was added to DOM after content, not before)
+
+**Automated testing complements but does not replace manual testing.** Automated tools catch ~30-40% of accessibility issues (missing alt text, missing labels, contrast violations). Manual screen reader testing catches the remaining 60-70% (poor announcements, confusing navigation, broken focus management).
+
+## Source
+
+https://www.w3.org/WAI/test-evaluate/
+
+## Process
+
+1. Read the instructions and examples in this document.
+2. Apply the patterns to your implementation, adapting to your specific context.
+3. Verify your implementation against the details and edge cases listed above.
+
+## Harness Integration
+
+- **Type:** knowledge — this skill is a reference document, not a procedural workflow.
+- **No tools or state** — consumed as context by other skills and agents.
+
+## Success Criteria
+
+- The patterns described in this document are applied correctly in the implementation.
+- Edge cases and anti-patterns listed in this document are avoided.

package/dist/agents/skills/codex/a11y-screen-reader-testing/skill.yaml

@@ -0,0 +1,43 @@
+name: a11y-screen-reader-testing
+version: '1.0.0'
+description: Test web applications with screen readers to verify accessible user experience
+cognitive_mode: advisory-guide
+type: knowledge
+tier: 3
+triggers:
+  - manual
+platforms:
+  - claude-code
+  - gemini-cli
+  - cursor
+  - codex
+tools: []
+paths:
+  - '**/*.tsx'
+  - '**/*.html'
+  - '**/*.vue'
+  - '**/*.svelte'
+  - '**/*.astro'
+related_skills:
+  - a11y-aria-patterns
+  - a11y-semantic-html
+  - a11y-testing-automation
+stack_signals:
+  - accessibility
+  - react
+  - typescript
+keywords:
+  - screen-reader
+  - voiceover
+  - nvda
+  - jaws
+  - narrator
+  - talkback
+  - testing
+metadata:
+  author: community
+  upstream: w3.org/WAI/test-evaluate
+state:
+  persistent: false
+  files: []
+depends_on: []