- Home
- Skills
- Frontend & Web Apps
- A11y Doctor — WCAG 2.2 Accessibility Audit & Fix
More screenshots
Works with the AI Agents you already use
A11y Doctor — WCAG 2.2 Accessibility Audit & Fix
Audit, prioritize, and fix WCAG 2.2 A/AA accessibility issues in web and mobile codebases.
$9.99
A11y Doctor — WCAG 2.2 Accessibility Audit & Fix
Example session with this skill installed
Check the product card in my React app for accessibility problems. Don't change anything yet—show me the most important issues first, explain what should be fixed, and tell me what I still need to test manually.
- Read your context and instructions
- Compiled the a11y doctor
- Generated the app files
Accessibility audit: Product card
I found the product card at
src/components/ProductCard.tsxand audited it against WCAG 2.2 Level A and AA. I found
9 open findings across 7 rules:
5 critical, 2 serious, and 2 minor. The built-in static scanner was used; this project does not currently have an accessibility ESLint plugin installed.
Priority Finding WCAG 2.2 Location Recommended change Critical Two images have no altattribute1.1.1 Lines 11 and 13 Confirm each image's purpose. Add meaningful alt text when informative, or alt=""when decorative; do not guess.Critical Two click handlers are attached to non-keyboard-operable elements 2.1.1, 4.1.2 Lines 15 and 26 Prefer native <button type="button">elements. If native HTML is not possible, supply the complete role, focus, Enter, and Space behavior.Critical An icon-only button has no accessible name 4.1.2 Line 17 Add visible text or a specific aria-label, and hide the decorative icon from assistive technology.Serious An anchor without a real destination is being used as a button 2.1.1, 4.1.2 Line 19 Use a button for the action; reserve links for navigation. Serious tabIndex={2}overrides the natural focus sequence2.4.3 Line 23 Remove the positive tabindex and correct the DOM order instead. Minor “Read more” is ambiguous outside its visual context 2.4.4 Line 21 Include the product name in the link text or add visually hidden context. Minor A backdrop closes on pointer click 2.1.1 Line 35 Verify that Escape also closes the dialog, a close control is keyboard reachable, and focus returns to the trigger. Recommended order
- Replace the clickable
divandliinteractions with semantic controls.- Give the icon-only button an accessible name.
- Confirm image meaning before adding alt text.
- Replace the action link and remove the positive tabindex.
- Improve the contextual link wording and verify dialog dismissal and focus return.
Testing gaps
The project currently has no automated axe test, accessibility lint plugin, documented screen-reader test plan, accessibility CI check, or A11y Doctor baseline. After applying fixes, re-run the same audit, add a component-level axe test, and consider a baseline-based CI gate for new serious findings.
Manual verification still required
Keyboard through every interactive element and confirm a logical focus order, visible focus, Escape behavior, and focus restoration. Test the component with VoiceOver and Safari and with NVDA and Firefox or Chrome. Confirm that image descriptions match their actual purpose and that the interaction remains understandable without color, pointer input, or visual context.
ProductCard.tsx
ZIP · project files
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Make accessibility part of every code change.
A11y Doctor helps your coding agent audit code for WCAG 2.2 Level A and AA issues, apply targeted fixes, and re-scan the result. Use it on a component, changed files in a pull request, a project folder, a running page, or Storybook.
Designed for: web projects using React, Next.js, Angular, or Vue, plus accessibility guidance for React Native, iOS, Android, and Flutter.
What it does
- Audits: combines built-in static checks with supported tools already available in your project, such as ESLint accessibility plugins and axe-core. For running interfaces, it can guide browser and keyboard checks.
- Prioritizes: reports findings by severity and groups related issues so you can address shared components or design tokens efficiently.
- Fixes: prefers native HTML and uses established ARIA patterns when needed. It makes focused changes, avoids suppressing rules, and does not invent alt text when author input is required.
- Verifies: re-runs applicable checks after changes and provides manual test steps for behavior that automation cannot reliably validate, including screen-reader and keyboard interaction.
- Supports prevention: can help configure a baseline-based CI check so newly introduced findings can be caught without requiring all existing findings to be fixed at once.
Included: an audit engine, guidance covering the 55 WCAG 2.2 Level A and AA success criteria, framework and native-platform references, ARIA widget patterns, and test templates for tools such as jest-axe, Playwright, and Guidepup. Automated coverage varies by criterion and project setup; some accessibility requirements always need human review.
Requirements: Node.js 20+ and Git for the packaged audit scripts. Some checks require a project-provided tool or a local development/Storybook URL. A11y Doctor does not send project code to a separate A11y Doctor service.
Try asking: “Audit my checkout page for accessibility and fix what you can,” or “Check the changed files in my pull request for accessibility issues.”
Use A11y Doctor through an Agensi-supported, connected coding agent. When the creator publishes an approved update, Agensi automatically loads that version the next time the skill is used.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Frontend & Web Apps.
Requires Node.js 20+. Cannot automate subjective checks like alt text meaningfulness or logical reading order. Manual review required for custom widgets.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
- Software
Node.js@axe-core/clipa11ylighthousePlaywright@guidepup/setup
Review Agensi's auto-detected permissions and use the following declarations.