wcag-audit-patterns

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs auditing arbitrary websites and shows runtime flows that load and analyze web pages (Automated Testing: Playwright's page.goto('/') and axe.run(document); CLI examples: npx @axe-core/cli https://example.com, pa11y https://example.com, lighthouse https://example.com), so it ingests untrusted public web content which could contain instructions that influence tool use or decisions.