mqttnet

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

• Third-party content exposure detected (high risk: 0.80). The skill subscribes to MQTT topics (e.g., vandaemon/leddimmer/+/status and vandaemon/leddimmer/{deviceId}/config) and deserializes/acts on device-published JSON and state messages as part of its Device Discovery and Message Handling workflows (references/workflows.md and references/patterns.md), which are untrusted third‑party messages that directly drive registration, state changes, and subsequent actions.