survey

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly requires web-enabled research and ingestion of public sources (e.g., "Search broadly in English", "use generic web-capable agents", "direct user voices from communities", and "include source links") so the agent will fetch and interpret untrusted third-party content from the open web that can materially influence its outputs and actions.