developing-genkit-python

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

• Third-party content exposure detected (medium risk: 0.30). In the “Agent HTTP” FastAPI runtime, the serve_agent(agent) route accepts requests from remote clients and feeds the posted chat input (e.g., via chat.send(...)) into the agent’s turn, meaning outsider-authored free text can reach the LLM without first selecting a specific item.