agentsclimarketplace
Sign in

A11y

Skill Syo-M/fable-frontend-skills/.claude/skills/a11y

Accessibility conventions — semantic HTML, keyboard support, focus management, ARIA, testing. Use when building UI components, forms, dialogs, navigation, or fixing a11y violations. 日本語の依頼例:「アクセシビリティ対応して」「キーボード操作できるように」「フォーカス管理」「スクリーンリーダー対応」「コントラスト」。From its SKILL.md

Install
npx -y skills add Syo-M/fable-frontend-skills --skill a11y

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.3 KB, 716 tokens by cl100k_base, as published. Nobody here has run it

Accessibility

Target: WCAG 2.2 AA. A11y is a build-time concern, not an audit-time patch.

Semantic HTML first

  • Use the element that does the job: <button> for actions, <a href> for navigation, <label>+input for forms, <nav>/<main>/<header> landmarks, real heading hierarchy (one <h1>, no level skipping).
  • A <div onClick> is never acceptable for an action — it has no keyboard support, no role, no focus. If a design demands a non-button look, style a <button>.
  • First rule of ARIA: don't use ARIA when an HTML element provides the semantics. ARIA adds promises (keyboard behavior) that you must then implement by hand.

Keyboard

  • Everything operable by mouse is operable by keyboard: Tab reaches it, Enter/Space activates it, Escape dismisses overlays, arrow keys move within composite widgets (menus, tabs, listboxes).
  • Visible focus indicator always — never outline: none without an equal-or-better :focus-visible style (use a token, see css-modules).
  • DOM order = visual order = tab order. Don't fix layout problems with tabindex > 0 (banned).

Focus management (SPAs & overlays)

  • Dialogs/drawers: focus moves in on open, is trapped while open, returns to the trigger on close. Prefer native <dialog> or a headless library (React Aria, Radix) over hand-rolling — focus traps are notoriously buggy.
  • Client-side route change: move focus to the new page's heading (or announce via live region) — silent navigation strands screen-reader users.
  • Content removal: if the focused element disappears, move focus somewhere sensible, not <body>.

Forms

  • Every input has a programmatic label (<label htmlFor>); placeholder is not a label.
  • Errors: associate with the field via aria-describedby, set aria-invalid, and on submit failure move focus to the first error or an error summary. Error text says how to fix, not just "invalid".
  • Don't disable the submit button as the only validation feedback — disabled buttons explain nothing.

Content & visuals

  • Contrast: text ≥ 4.5:1 (3:1 for large text), UI component boundaries ≥ 3:1. Check tokens once in tokens.css, inherit everywhere.
  • Color never the sole signal — pair with text/icon (error = red AND icon + message).
  • Meaningful images need alt describing function; decorative ones alt="". Icon-only buttons need aria-label.
  • Async updates (toasts, validation, loading→loaded) announce via aria-live="polite" regions that exist in the DOM before the update.
  • Respect prefers-reduced-motion for any significant animation.

Testing

  • Automated catches ~30–40%: axe in Storybook (a11y addon, failing) + @axe-core/playwright on key pages. Zero serious/critical violations.
  • The rest is manual: keyboard-walk every new interactive component (Tab/Enter/Escape/arrows) as part of done.
  • Writing tests with getByRole(…, { name }) (per testing-vitest) doubles as an a11y check — if the role query can't find it, assistive tech can't either.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 1 of the 12 instructions most accessibility skills give in 716 tokens

Counted across 360 of the 378 authors here whose files we hold, read 2026-09-06

  • Respect prefers-reduced-motionin 49 of 360, across 41 files
  • Add alt text to every imagein 47 of 360, across 40 files
  • Make every interactive element keyboard operablein 46 of 360, across 37 files
  • Keep focus indicators visiblehere, and in 42 of 360, across 37 files
  • Associate every form input with a labelin 42 of 360, across 35 files
  • Ensure 4.5:1 contrast for normal textin 38 of 360, across 32 files
  • Trap focus in modals and restore it on closein 37 of 360, across 25 files
  • Prefer native HTML elements over ARIAin 37 of 360, across 33 files
  • Add aria-label to icon-only buttonsin 24 of 360, across 13 files
  • Announce dynamic content updates with aria-livein 24 of 360, across 16 files
  • Use ARIA only when native HTML is insufficientin 23 of 360, across 15 files
  • Meet WCAG AA contrast ratiosin 23 of 360

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.