NYC

iot-developer

Fail

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: HIGHDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • DATA_EXFILTRATION (HIGH): The skill directs users to use 'broker.hivemq.com', which is a public, unauthenticated MQTT broker. Any data published to topics like 'home/temperature' or 'home/humidity' is visible to anyone on the internet, violating data privacy for smart home or industrial contexts.
  • COMMAND_EXECUTION (HIGH): The open nature of the public broker allows any third party to publish messages to control topics (e.g., 'home/light/1'). This enables unauthorized actors to trigger physical side effects on devices managed by the skill, such as toggling lights or adjusting thermostats.
  • EXTERNAL_DOWNLOADS (LOW): The skill requires the installation of the 'mqtt' Node.js package without specifying a version, which is a sub-optimal security practice for supply chain integrity.
  • Malicious URL Detection (MEDIUM): Automated scanners flagged the HiveMQ broker URLs as malicious/blacklisted, which aligns with their use in insecure or unauthenticated IoT implementations.
Recommendations
  • AI detected serious security threats
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 09:21 PM