foundry-agent

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the agent’s LLM context via the web_search tool (agent.yml → tools entry with type: web_search and function_name: search_web), which at runtime fetches public web content and supplies it to the model for answering.