search

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to fetch and ingest untrusted public web and platform content (e.g., Tavily's /crawl and /map, Exa's /contents, and SearchAPI engines such as youtube_comments, youtube_transcripts, social media profiles and ad libraries), which the agent is expected to read and use to produce sourced answers and drive follow-up actions, meeting the criteria for indirect prompt injection risk.