accessibility-audit
Audit and fix accessibility (a11y) — WCAG 2.2 AA compliance, screen-reader flows, keyboard navigation, and ARIA correctness for web apps.
Skill by xanstomper
Chalk Labs
Need help deploying accessibility-audit?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/xanstomper-accessibility-audit
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design, security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/accessibility-audit
What's inside
- Skill name
accessibility-audit- Description
- Audit and fix accessibility (a11y) — WCAG 2.2 AA compliance, screen-reader flows, keyboard navigation, and ARIA correctness for web apps. Use when checking a UI for accessibility, fixing a11y bugs, or preparing for compliance review.
- License
- MIT
- Instructions
- 34 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Accessibility Audit Procedure Accessibility is correct semantics, not ARIA sprinkles. The stack: semantic HTML > ARIA > nothing. The test: unplug the mouse, unplug the eyes. ## 1. Keyboard pass (the cheapest audit) - Tab through the whole flow: every interactive element reachable, visible focus (never `outline: none` without replacement), logical order, no traps. - Modal traps: focus moves IN on open, returns to the trigger on close, `Esc` closes, tab cycles INSIDE the modal only. - Custom widgets (tabs, dropdowns, carousels) implement their keyboard contract: arrow keys per WAI-ARIA authoring practices — arrows for tabs/options, Space/Enter for activation. ## 2. Screen-reader pass (NVDA free on Windows; VoiceOver on macOS/iOS) - Landmarks: one `<main>`, header/nav/footer as elements, skip-link first. Headings: hierarchical (h1 unique, no level skips), nothing conveyed by "bold big text". - Name-role-value: every control announces what it IS and DOES. Icon-only buttons need `aria-label`; toggles announce state (`aria-pressed`/`aria-expanded`). - Live regions: async updates (toasts, search counts, loading states) announced via `aria-live="polite"` — silent UI updates are invisible to SR users. - Forms: every input has a programmatic label (`for`/`id`, not placeholder-only); errors linked via `aria-describedby` and announced; required marked in text, not color alone. ## 3. Automated floor (catches ~30-40%; the rest is manual) - axe-core in CI (Playwright integration) with zero violations on key pages; Lighthouse a11y section >= 95. - But: axe cannot tell you tab ORDER is wrong, focus is lost on route change, or the flow is unusable. Manual passes above are the real audit. ## 4. Visual standards - Contrast: 4.5:1 body, 3:1 for >=24px text and UI boundaries/focus indicators. Zoom to 200% and mobile: no loss of function, no horizontal scroll for single-column prose. - Motion: `prefers-reduced-motion` honored; nothing flashes >3x/second (seizure threshold). - Color-blindness: run through deuteranopia/protanopia simulation — never color-alone status; pair with icon/text. ## 5. Fix priorities (impact-ordered) - Blockers first: unreachable flows, unlabeled forms, broken focus order. Then: SR announcements, contrast, target sizes (24x24 CSS px minimum). - Fix at the component layer (design system), not per-page — a fix in the button component fixes every page at once. ## Pitfalls - `role="button"` on a div instead of using `<button>`: you've rebuilt HTML badly. Native elements carry keyboard/SR behavior for free. - aria-label overuse: it OVERRIDES content — `aria-label="Close"` on a button containing "Close ×" hides the visible text from SR users. - Testing with the dev tools' a11y tree only: it lies by omission (not all properties shown). A real screen reader is the oracle.
From xanstomper/mochi, MIT.
More from xanstomper/mochi
- 0.587score
api-backwards-compatibility
xanstomper
Change APIs without breaking existing clients — deprecation windows, additive changes, and safe removal.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-design
xanstomper
Design public APIs and CLIs — naming, ergonomics, versioning, error surfaces, and backward-compat discipline.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-integration
xanstomper
Integrate third-party APIs correctly — auth patterns, retry/backoff, pagination, rate limits, idempotency, and contract testing.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-security-hardening
xanstomper
Secure web APIs and backends — authz enforcement, injection prevention, rate limiting, secret handling, and the OWASP-class checks that block real attacks.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-versioning
xanstomper
Evolve APIs without breaking clients — additive change first, deprecation windows, compatibility shims, and migration paths.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
async-concurrency
xanstomper
Design correct async and concurrent code — bounded parallelism, backpressure, cancellation, and deadlock avoidance.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
auth-implementations
xanstomper
Implement authentication correctly — session/JWT/OAuth/SSC flows, password handling, MFA, and the failure modes that become breaches.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
benchmark-methodology
xanstomper
Honest benchmarking with TTFF/TTFI/PSS probes — cold start, token throughput, prompt-to-first-token.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
browser-automation
xanstomper
Automate real browsers for testing and scraping — Playwright-first flows, selector strategy, waiting discipline, and anti-flake patterns.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
caching-strategy
xanstomper
Design caching layers — HTTP caching, CDN, application memory caches, invalidation patterns, and cache-coherence reasoning.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
chaos-engineering
xanstomper
Prove a system survives failure by injecting it deliberately — game days, fault injection, and steady-state hypotheses.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
checkpoint-semantics
xanstomper
Git-backed checkpoints and rollback composition — naming, nesting, conflicts, when to auto-commit.
SkillAFreshness: A★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 2
- Open issues
- 6
- Last commit (this folder)
- Oct 12, 2026 (19 days ago)
- Last release
- Oct 23, 2026 (9 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/xanstomper/mochi
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from xanstomper
- 0.587score
api-design
xanstomper
Design public APIs and CLIs — naming, ergonomics, versioning, error surfaces, and backward-compat discipline.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-integration
xanstomper
Integrate third-party APIs correctly — auth patterns, retry/backoff, pagination, rate limits, idempotency, and contract testing.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-security-hardening
xanstomper
Secure web APIs and backends — authz enforcement, injection prevention, rate limiting, secret handling, and the OWASP-class checks that block real attacks.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-versioning
xanstomper
Evolve APIs without breaking clients — additive change first, deprecation windows, compatibility shims, and migration paths.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
async-concurrency
xanstomper
Design correct async and concurrent code — bounded parallelism, backpressure, cancellation, and deadlock avoidance.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
api-backwards-compatibility
xanstomper
Change APIs without breaking existing clients — deprecation windows, additive changes, and safe removal.
SkillAFreshness: A★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get accessibility-audit running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
accessibility-audit skill. Agent Mart. https://agentmart.dev/skills/xanstomper-accessibility-audit (last updated 2026-10-24)