multi-search-engine

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md and references (e.g., SKILL.md "Quick Examples" and references/international-search.md) explicitly instruct the agent to perform web_fetch on public search engines and sites (Google, DuckDuckGo, Brave, WolframAlpha, etc.), meaning the agent will fetch and read untrusted, user-generated public web content that can influence subsequent decisions or tool use.