heartbeat

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md shows that it uses Membrane to run actions and proxy arbitrary requests to a Heartbeat instance (e.g., "membrane action run" and "membrane request ..."), including listing/reading user-generated resources like threads and documents, which the agent is expected to fetch and interpret as part of its workflow.