agent-memory-systems

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text could enter the LLM context via the “Background memory formation” runtime path: it loads a conversation (load_conversation(thread_id)) and then interpolates that conversation text directly into an LLM prompt (llm.invoke(...)), where the conversation content may include messages authored by non-operating users.