Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add varunk130/ai-ux-skill-library --skill ai-accessibility-auditgit clone --depth 1 https://github.com/varunk130/ai-ux-skill-libraryWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/varunk130/ai-ux-skill-library/ai-accessibility-audit)<a href="https://agentmods.dev/skills/varunk130/ai-ux-skill-library/ai-accessibility-audit"><img src="https://agentmods.dev/badge/skills/varunk130/ai-ux-skill-library/ai-accessibility-audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/varunk130/ai-ux-skill-library/ai-accessibility-audit"><img src="https://agentmods.dev/badge/skills/varunk130/ai-ux-skill-library/ai-accessibility-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00088 | $0.02322 |
| Opus 5 | $0.00044 | $0.01161 |
| Sonnet 5 | $0.00018 | $0.00464 |
| Haiku 4.5 | $0.00009 | $0.00232 |
Grade A, and why
ai-accessibility-audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Accessibility Audit
Catch accessibility regressions in AI product surfaces before they ship - chat interfaces, agent dashboards, generated-content panels, streaming output regions, and consent / safety UI. WCAG 2.2 Level AA is the bar; the framework below operationalizes it for AI-specific patterns that the spec doesn't natively cover.
Core Principle
Accessibility is a property of the experience, not just the markup. An AI product can pass an axe-core scan and still be unusable for a screen-reader user if streaming output isn't announced, consent UI traps focus, or "regenerate" controls are unreachable from the keyboard. This skill audits both the static layer (DOM, contrast, semantics) and the AI-specific dynamic layer (live regions, latency, motion).
The CLEAR Audit Framework
| Letter | Dimension | The Question |
|---|---|---|
| C | Contrast & Color | Does every text/icon/state meet 4.5:1 (or 3:1 for ≥18pt or non-text UI)? Is color the only signal anywhere? |
| L | Logic of Focus | Can a keyboard-only user reach every action in a sensible order, with a visible focus indicator and no traps? |
| E | Equivalents for Non-Text | Do images, charts, generated visuals, and audio have meaningful alternatives - including AI-generated content? |
| A | Announcements & Live Regions | Are streaming outputs, status changes, and AI errors announced to assistive tech without spamming? |
| R | Respect for User Preferences | Does the product honor prefers-reduced-motion, prefers-color-scheme, OS-level font scaling, and reduced-data hints? |
WCAG 2.2 AA Coverage Matrix
The skill produces findings against the success criteria most commonly violated in AI product surfaces:
| Success Criterion | What Gets Checked | AI-Specific Gotchas |
|---|---|---|
| 1.1.1 Non-text Content | Alt text on every meaningful image | AI-generated images often ship with empty alt; charts rendered as <canvas> lack text equivalents |
| 1.3.1 Info & Relationships | Headings, lists, landmarks, form labels | Markdown rendered from LLM output frequently produces orphan headings (h3 without h2) |
| 1.4.3 Contrast (Minimum) | 4.5:1 on all body text | "Subtle" UI states (disabled, secondary, "thinking…") often dip below threshold |
| 1.4.10 Reflow | No horizontal scroll at 320 CSS px | Long streaming code blocks and tables break this - needs horizontal-scroll container, not body scroll |
| 1.4.11 Non-text Contrast | 3:1 for UI components, focus indicators | Custom focus rings on dark backgrounds frequently fail |
| 1.4.13 Content on Hover/Focus | Tooltips dismissible, hoverable, persistent | "Why this answer?" popovers on AI citations often violate all three |
| 2.1.1 Keyboard | All functionality keyboard-operable | Send button, regenerate, stop-generation, copy-to-clipboard, model picker - audit each |
| 2.1.2 No Keyboard Trap | Focus can leave any component | Modal consent gates and citation popovers frequently trap |
| 2.4.3 Focus Order | Logical, predictable focus order | Streaming content insertion can shift focus or skip newly-rendered actions |
| 2.4.7 Focus Visible | Visible indicator on every focused element | outline: none with no replacement is the single most common failure |
| 2.4.11 Focus Not Obscured (2.2 new) | Focused element not hidden behind sticky UI | Sticky composer bars often cover the focused message above |
| 2.5.7 Dragging Movements (2.2 new) | Single-pointer alternative for drag | "Drag to reorder citations" patterns need keyboard alternative |
| 2.5.8 Target Size (2.2 new) | Interactive targets ≥24×24 CSS px | Inline action chips inside chat bubbles routinely fail |
| 3.2.6 Consistent Help (2.2 new) | Help mechanisms consistently located | "Report response" / feedback controls drift across surfaces |
| 3.3.7 Redundant Entry (2.2 new) | Don't make users re-type info in the same flow | Multi-step agent setup flows commonly violate |
| 3.3.8 Accessible Authentication (2.2 new) | No cognitive function test for auth | API-key paste flows that disallow paste fail this |
| 4.1.3 Status Messages | Use ARIA live regions for status without focus shift | Streaming AI output, "Thinking…", and error toasts must use aria-live="polite" (not assertive for streaming - it spams) |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 125 lines · 88 tokens per session scan A b5c8a6c3055a
ai-accessibility-audit is a skill published in the GitHub repository varunk130/ai-ux-skill-library (3 stars, last pushed 1mo ago), licensed MIT. It adds 88 tokens to every session and 2,322 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
brand-visual-language
A brand's visual tone — playful or serious, rounded or angular — should be consistent across all UI elements. Shape language in typography, border-radius, and iconography communicates personality before a single word is read. Use when establishing a design system, choosing icon libraries, setting border-radius tokens…
data-display-and-selection
Complex data deserves multiple view modes — grid, list, table — chosen by the user based on their task. Row and item selection should use large hit areas (the whole row or card, not just a checkbox). Selected state is communicated through a subtle background colour shift. Mass actions appear when items are selected.…
generate-ui-from-brand
Pipeline skill — turns a URL or DESIGN.md into a concrete UI structure with decisions already made. Extracts live design tokens, normalizes them into a semantic system, applies UX principles, and outputs an actionable UI spec. Use when building UI for an existing brand from scratch, auditing a design system, or…
information-architecture
In large applications, information architecture determines whether users can find, understand, and act on data. Naming matters. The UI should mirror the data model and signal how data can be transformed. Dangerous or irreversible changes always require a confirm dialog. Use when designing navigation, naming entities…
micro-interactions
Micro-interactions are small, purposeful animations and responses that reward the user and make the interface feel alive — an animated icon, a satisfying toggle, a subtle reveal. Borrowed from the natural world, they add delight without distraction. Use when designing interactive components, success states, toggles…
modal-and-overlay-patterns
Overlays — modals, drawers, bottom sheets, popovers — interrupt or augment the main flow. Each type has a different scope, blocking level, and appropriate use case. Use when designing dialogs, confirmation prompts, side panels, action sheets, or any UI element that appears above the main content layer.