domain-iot

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

• Third-party content exposure detected (high risk: 0.70). The skill subscribes to and processes MQTT messages from an external broker (see client.subscribe("devices/device-1/commands", ...) and eventloop.poll() in the run_mqtt code), which ingests untrusted, potentially user-generated third-party content that could carry indirect prompt-injection payloads.