rag-architecture

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md "Agentic RAG" section explicitly shows an agent tool list including Tool("search_web", "Search the web for current info", web_search"), which instructs the agent to fetch and act on open/public web content (untrusted third‑party sources) as part of its decision-making flow.