synapse

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow for this skill is NATS-based agent-to-agent messaging (register/discover/request/respond/emit/subscribe) where the agent LLM context is populated from message payload/text fields coming from other agents over the mesh (i.e., outsider-authored free-form request content), which can enable indirect prompt injection.