vibing-with-ntm

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow ingests outsider-authored free text via the “Mail / coordination” path: it reads Agent Mail inbox threads (which can contain messages authored by other agents/peers) into the orchestrator/LLM context through ntm --robot-snapshot/attention and the “Agent Mail inbox and threads — the thread is the source” grounding steps.