hass-declarative
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONNO_CODEPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation includes examples of high-privilege command execution using sudo and SSH for managing Home Assistant services and running maintenance scripts.
- [NO_CODE]: Core logic for the declarative management system, including the Python cleanup script and systemd service definitions, is referenced in the documentation but not provided in the skill source.
- [PROMPT_INJECTION]: The skill establishes a workflow that processes Nix configurations to perform destructive operations, representing an indirect prompt injection surface. 1. Ingestion points: Nix modules in the _domains/ directory; 2. Boundary markers: None identified in the described architecture; 3. Capability inventory: Modification of the Home Assistant entity registry and wiping of local YAML configuration files; 4. Sanitization: The process uses platform-based filtering to protect integration-managed entities.
Audit Metadata