weekly-performance-advisor

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow pulls the user’s LGM conversation messages/replies via LGM MCPs (e.g., get_conversation_messages for threads in get_conversations_to_reply / search_conversations), and those message bodies are free-form outsider-authored text that the agent must read and classify into the LLM context.