all-agentic-architectures

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly calls a web search tool (TavilySearchResults) to fetch internet search results and then injects those search_results directly into LLM prompts (e.g., the ReAct example, research_agent in the Multi-Agent section, and the meta-controller's research_specialist), exposing the agent to untrusted third-party content that can influence behavior.