home-assistant
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [CREDENTIALS_SAFE] (SAFE): The scripts utilize the
HA_TOKENenvironment variable to authenticate with Home Assistant, avoiding hardcoded secrets within the source code. - [NETWORK_SECURITY] (SAFE): All network communications (REST and WebSockets) are strictly directed at
homeassistant.local, a local network address. There is no evidence of data being transmitted to external or untrusted third-party domains. - [CODE_INTEGRITY] (SAFE): User inputs such as entity IDs, service names, and JSON data are passed to well-defined API functions. No instances of dynamic code execution (
eval,exec) or unsafe shell command interpolation were found. - [INDIRECT_PROMPT_INJECTION] (LOW): The skill ingests data from the user's Home Assistant instance (e.g., dashboard configurations and automation traces). This constitutes a standard indirect injection surface where content inside the HA instance could theoretically influence an agent's behavior, but the ingestion and display logic are implemented safely.
Audit Metadata