webapp-testing

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly lists "fetching content from external websites" and the Reconnaissance-then-action workflow (navigate, "Inspect rendered DOM", "Identify selectors", then "Execute actions") which require the agent to fetch and interpret arbitrary third-party web pages and then act on discovered content, enabling indirect prompt injection.