search

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly issues SerpApi queries to fetch search results from public websites and engines (see SKILL.md "API Pattern" and "Engine Selection"), and its required workflow instructs the agent to read and summarize those third-party search results, which could contain untrusted/user-generated instructions that influence subsequent actions.