Ux Accessibility Audit logo

Ux Accessibility Audit

Organization
Uxcel-Lab
ux-accessibility-audit

Run a structured WCAG 2.1 AA accessibility audit on a design, screen, component, or page. Produces a prioritized issue list — each finding cites the violated WCAG criterion, a severity rating (Critical/Major/Minor), and a concrete fix. Use when auditing for a11y compliance or as a validation step after generating UI.

Overview

PublisherUxcel-Lab
Repositoryproduct-skills
Skill nameux-accessibility-audit
Stars
35
Forks
8
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by Uxcel-Lab on GitHub. Read the source before you install it.

Installation

Install the Ux Accessibility Audit AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Uxcel-Lab/product-skills/tree/main/ux/audits/accessibility
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Uxcel-Lab/product-skills.git /tmp/product-skills
mkdir -p .claude/skills
cp -r /tmp/product-skills/ux/audits/accessibility .claude/skills/ux-accessibility-audit
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ux Accessibility Audit in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Ux Accessibility Audit on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Ux Accessibility Audit is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Accessibility Audit Skill

How this skill behaves (read first)

This is an evaluative skill — it reviews existing work against a standard (WCAG), it doesn't generate UI. Flagging is safe, so it auto-runs and is designed to be called as a validation step by generative skills as well as invoked directly.

It owns the accessibility concern in this repo. Generative skills hand off to it (e.g. ux-login-signup, ux-pricing), and the heuristics audit (ux-heuristics-audit) deliberately defers contrast / keyboard / screen-reader specifics here rather than duplicating them. When invoked, do the full a11y pass; don't assume another skill covered it.

Scope discipline. When invoked directly (the user named this audit), review only this concern — don't pull in sibling audits. It runs alongside other lenses only when an orchestrator (ux-design-review) or a generative skill calls it under docs/orchestration-policy.md, where it sits in Tier B — offered (and narrowed to usage when a design system is in play). Explicit scope always wins.


What this skill changes vs. default behavior

By default, accessibility feedback tends to be partial and vague — a few obvious call-outs ("add alt text", "check contrast") with no standard, no severity, and no systematic sweep. This skill forces four things:

  1. Every finding cites the WCAG criterion and level it violates — not a vague suggestion.
  2. Every finding gets a severity based on how completely it blocks access — so the list is prioritized.
  3. Every finding has a concrete fix — the specific markup/design change.
  4. The audit is systematic — it sweeps all six categories below, so issues outside the obvious ones (focus order, heading hierarchy, table semantics, motion) actually get caught.

WCAG foundations

POUR principles — every audit item maps to one of these four:

  • Perceivable — content must be detectable via at least one sense (sight, hearing, etc.)
  • Operable — all interactions must work without a mouse (keyboard, switch, voice)
  • Understandable — content and UI must be clear and predictable
  • Robust — content must work across browsers, devices, and assistive technologies

Compliance levels:

  • Level A — bare minimum; missing these creates complete barriers (e.g. no alt text, no keyboard access).
  • Level AA — the legal standard most organizations must meet; the default target for every audit.
  • Level AAA — aspirational; apply selectively where it adds high value.

Default audit target: WCAG 2.1 Level AA.


Audit checklist

Work through every category — a partial sweep is the main way real issues get missed. Flag each violation with WCAG criterion, severity (see model below), and a fix.

1. Color & Contrast

CheckRequirementNotes
Normal text vs background≥ 4.5:1 (AA) / 7:1 (AAA)Applies to all text in UI: buttons, labels, cards, nav
Large text (≥ 18pt or ≥ 14pt bold)≥ 3:1
Icons (informational or interactive)≥ 3:1Decorative icons exempt
Input borders vs surrounding background≥ 3:1Including focus indicators
Text over images≥ 4.5:1 between text and the portion of image behind itUse semi-transparent overlay if needed
Selection statesSame ratios as regular text
Disabled elementsExempt from contrast rulesMust still be recognizable as inactive
Logo textExemptBut good practice to maintain contrast anyway
Color as sole differentiator❌ NeverAlways pair color with icon, pattern, label, or text
Stark contrast (pure black on white)DiscouragedUse #333 on #F8F8F8 — exceeds AA while reducing eye strain

Don't rely on color alone: error states need an icon + text, not just red. Charts need patterns or labels, not just color.


2. Typography & Text

CheckRequirement
Base body font size≥ 16px
Line length45–75 characters (including spaces)
Text alignmentFollows reading direction (LTR → left-align; RTL → right-align)
Headings visually distinctLarger size, bold weight, or color difference from body text
One <h1> per pagePage title only
Heading hierarchyNo skipped levels (h1 → h2 → h3, never h1 → h3)
Semantic heading tagsUse <h1>–<h6>, not styled <div> or <p>
LanguagePlain language, secondary-education reading level
Idioms & figures of speechAvoid — confuse screen readers and non-native speakers
TypefaceClear letterform distinctions (0 vs O, 1 vs l vs I); large x-height

3. Forms

CheckRequirement
Keyboard accessEvery field, dropdown, checkbox, radio, button navigable via Tab / Shift+Tab / Enter / arrows
Focus indicatorsVisible on every focused element — never remove browser default without replacing
LayoutSingle-column preferred; multi-column increases cognitive load and screen-reader confusion
Input boundariesVisible borders or backgrounds — never borderless inputs
LabelsOutside the field, not placeholder-only; min 16px; 4.5:1 contrast
Placeholder textSupplemental only — disappears on input, so never the sole label
AutocompleteImplement for name, email, phone, address, payment (HTML autocomplete attribute)
Error statesMultiple cues: icon + bold border + field-specific message. Never color alone
Error placementAdjacent to the offending field + aria-describedby linking field to error
Error summaryAt top of form on submit failure, with anchor links to each problem field
AuthenticationNo text-based CAPTCHA — use magic links, OTP, passkeys, or biometrics. Support paste in password fields
Mobile one-handed useCenter CTAs in lower-middle zone; avoid right-edge-only button placement

4. Links & Navigation

CheckRequirement
Link labelsDescriptive, specific — "Download pricing guide" not "Click here" or "Read more"
New-window linksWarn users in link text or aria-label: "opens in new tab"
File linksState type + size: "Annual Report (PDF, 2.5 MB)"
Raw URLs as link text❌ Never — screen readers read character by character
Link recognizabilityUnderline + color (not color alone)
Focus stateVisible, distinctive focus ring on every link and interactive element
href attributeRequired on all <a> tags — <div> with click handler is not a link
Voice/toneLink text should match the product's voice, but always prioritize clarity
Help linksConsistently placed (footer, header utility nav, or persistent help icon)

5. Images & Multimedia

CheckRequirement
Informational imagesalt attribute with meaningful description
Decorative imagesalt="" (null) — tells screen readers to skip
Images containing textAlt text = the exact text in the image
Autoplay❌ Disabled by default — user must initiate playback
Media controlsPlay/pause/stop/rewind/volume/mute/captions — keyboard accessible, sufficient size and contrast
Video captionsRequired for all pre-recorded video with audio
Audio transcriptsRequired for all audio-only content
Flashing/strobing content❌ No content flashing > 3 Hz or red flashing at any rate
Animations (GIF, SVG, CSS)No more than 5 light-dark stripe pairs; provide a way to pause/stop
IconsPair with text labels; informational icons need aria-label or title
Swipe gesturesSupport both directions or provide an alternate tap target

6. Tables & Lists

CheckRequirement
Data tablesUse <table> element — never use tables for layout
Table caption<caption> immediately after <table> opening tag
Header cells<th> with scope="col" or scope="row" — never empty
Complex tablesAvoid merged cells and multi-level headers; split into simpler tables if needed
Cell widthsUse % not px — allows browser reflow; never set cell height manually
ListsUse <ul>, <ol>, or <dl> appropriately — not styled <div> rows
Unordered listsFor items where order doesn't matter
Ordered listsFor sequential steps, rankings, instructions
Description lists<dl> / <dt> / <dd> for glossaries, FAQs, key-value metadata

Severity rating

Rate each finding by how completely it blocks access and which WCAG level it fails:

SeverityMeaning
CriticalBlocks access entirely for some users. Usually a Level A failure — no keyboard access, no alt text on informational images, missing captions, a keyboard trap. Fix before ship.
MajorSignificant friction; users can sometimes work around it but shouldn't have to. Usually a Level AA failure — contrast below 4.5:1, placeholder-as-label, color-only error states, missing focus indicators.
MinorNoticeable but low-impact, or a best-practice / AAA improvement — slightly short line length, stark pure-black text, a non-ideal but functional label.

When in doubt, rate by the user impact (does it block a task or merely add friction?), not by how easy the fix is.


Audit output format

## Accessibility Audit — [Component / Screen Name]
Context: [platform · WCAG target (default AA) · design or code]

### Critical (blocks access entirely)
- [Issue]: [what's wrong] → [fix] — WCAG [X.X.X]

### Major (significant friction)
- [Issue]: [what's wrong] → [fix] — WCAG [X.X.X]

### Minor (best practice / AAA)
- [Issue]: [what's wrong] → [fix] — WCAG [X.X.X]

### Passed ✓
- [what's already working]

Always end with a one-line summary stating the overall WCAG AA compliance status and the top fixes that would get it there.


How to run a good audit

  1. Establish context. Platform (web/mobile/native), target level (AA by default — only go AAA if asked), and whether you're auditing a design or live code (some checks, like keyboard order and aria-*, only apply to code).
  2. Sweep all six categories. The value of this skill over a default review is completeness — don't stop at contrast and alt text.
  3. Rate by barrier, not by effort. A one-line alt fix can still be Critical.
  4. Be specific in fixes. Give the exact change (markup, attribute, ratio, or design adjustment), not "make it accessible."
  5. Note tooling limits. Automated scanners catch only ~30–40% of issues — manual review of focus order, alt-text quality, and reading order still matters.

Recommended tools

For automated scanning (catches ~30–40% of issues):

  • WAVE (browser extension) — overlays visual indicators, references violated WCAG criteria
  • axe DevTools (Chrome extension) — accurate, dev-friendly; powered by axe-core
  • Google Lighthouse (Chrome DevTools → Lighthouse tab) — accessibility + SEO + performance in one pass

For contrast checking in design:

  • Able (Figma plugin) — contrast checker + color-blindness simulator
  • Stark (Figma / Sketch / Adobe XD) — contrast, vision simulator, focus order, alt-text annotations
  • WCAG Color Contrast Checker (Chrome extension) — real-time ratio checking in browser

For screen-reader testing:

  • VoiceOver (macOS/iOS, built-in) — navigate with VO + arrow keys
  • NV Access / NVDA (Windows, free) — most widely used Windows screen reader

For seizure risk:

  • PEAT (downloadable) — analyzes video/animation for photosensitive epilepsy triggers
  • Harding Test — online paid service for broadcast/video content

Common do/don't patterns

❌ Don't✅ Do
Use red text alone for errorsRed border + warning icon + descriptive message
<a>Click here</a><a>Download the onboarding guide (PDF, 1.2 MB)</a>
<img src="chart.png"> (no alt)<img src="chart.png" alt="Bar chart showing Q3 revenue by region">
Decorative image with alt text<img src="bg-swoosh.svg" alt="">
Pure black #000 on pure white #FFFDark grey #333 on off-white #F8F8F8
Placeholder as the only labelPersistent label above the field + optional placeholder
Jump from <h1> to <h3>Always use <h1> → <h2> → <h3> in order
Table for page layoutCSS Grid or Flexbox for layout; <table> for data only
Autoplay videoControls visible at top; user initiates playback
Color-only chart legendAdd patterns, shapes, or data labels alongside color
Report only the obvious issuesSweep all six categories; rate by barrier; cite the criterion

Source lessons (Uxcel)

Frequently asked questions

What does the Ux Accessibility Audit AI skill do?

Run a structured WCAG 2.1 AA accessibility audit on a design, screen, component, or page. Produces a prioritized issue list — each finding cites the violated WCAG criterion, a severity rating (Critical/Major/Minor), and a concrete fix. Use when auditing for a11y compliance or as a validation step after generating UI.

Why use Ux Accessibility Audit on TypingMind?

Because you install it once and use it with any model. Ux Accessibility Audit is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Ux Accessibility Audit in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Uxcel-Lab/product-skills/tree/main/ux/audits/accessibility. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Ux Accessibility Audit?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Ux Accessibility Audit?

As many as you like. As long as a model supports skills, you can use Ux Accessibility Audit with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Ux Accessibility Audit AI skill free?

Yes. It is published on GitHub by Uxcel-Lab under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇