claim-investigation

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs investigating claims from social media and viral content and directs the agent to search and cross-reference public news sites, social media posts, and other open web sources (see "investigate claims from social media and other sources", "Search Strategy", and Phase 4 source types), so the agent will ingest and interpret untrusted third‑party user-generated content.