qe-accessibility-testing

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Agent-Driven Accessibility" workflow and test examples (e.g., the Task with url: 'https://example.com/checkout' and page.goto('/') / page.goto('/checkout') in the axe-core and Playwright snippets) show the agent visiting and analyzing web pages, meaning it fetches and interprets arbitrary third-party web content as part of its testing flow, which could allow indirect prompt-injection from untrusted pages.