Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Runs accessibility scans on live web pages using the axe DevTools MCP Server via Docker. Returns violations with rule IDs, impact levels, and affected selectors, then fetches AI-generated fix guidance per violation. Also applies static accessibility best practices when no live page is available.
Combines automated scanning and targeted remediation in one workflow, so developers get specific code fixes rather than just a list of rule names.
When to use it
- Scanning a locally running app for WCAG violations before a PR
- Getting a code fix for a specific broken ARIA attribute
- Verifying a form component has properly associated labels
- Checking color contrast and heading hierarchy in a new page
- Re-running a scan after fixes to confirm zero remaining violations
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: check my localhost:3000 checkout form for accessibility issues
- 1Run analyze against http://localhost:3000/checkout
- 2Parse violations from the JSON response, grouping by rule ID
- 3Call remediate for each unique violation with its ruleId, element HTML, and description
- 4Apply the returned code_fix patches to the source files
- 5Re-run analyze to confirm zero violations remain
Checkout form passes axe scan with no violations; label associations, focus management, and contrast issues resolved
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.