ha-error-checking

Pass

Audited by Gen Agent Trust Hub on Feb 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill behavior is consistent with its stated purpose of debugging Home Assistant configurations. No prompt injection or obfuscation was found.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: Sensitive credentials like the Home Assistant Long Lived Token are retrieved from environment variables (HA_LONG_LIVED_TOKEN), which is a secure practice. Network operations are directed to a local IP address and are restricted to the Home Assistant WebSocket API.
  • [EXTERNAL_DOWNLOADS]: The skill contains logic to install HACS cards using the hacs/repository/download WebSocket command. This is a legitimate administrative function within the context of Home Assistant management and does not involve executing untrusted remote code on the local system.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests system logs and dashboard configurations from the Home Assistant instance. While these represent untrusted data sources, the skill uses the data solely for validation and reporting. It does not interpolate this data into agent instructions or execute it as code, effectively mitigating injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 24, 2026, 05:23 PM