iot-expert

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides educational content and code templates for IoT systems development. The instructions and code snippets are well-structured and do not contain malicious patterns.
  • [CREDENTIALS_UNSAFE]: The skill avoids hardcoding credentials. It explicitly provides a recommendation to use a separate secrets.h file for WiFi credentials and to exclude it from version control using .gitignore.
  • [DATA_EXFILTRATION]: Network operations in the provided code snippets (MQTT and HTTP placeholders) are limited to telemetry data and use example domains (broker.example.com, updates.example.com). No exfiltration of sensitive local data was detected.
  • [REMOTE_CODE_EXECUTION]: No remote code execution patterns, such as piping network content to a shell, were found. The skill uses standard, well-known libraries for its functionality.
  • [COMMAND_EXECUTION]: The skill does not include any commands that execute shell scripts or system-level operations that could lead to unauthorized access or command injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:16 PM
Security Audit — agent-trust-hub — iot-expert