Security audit
home-assistant-master
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only Home Assistant operations skill that clearly defaults to read-only diagnostics and requires explicit confirmation for device or platform changes.
Install only in a private, trusted agent environment connected to the correct Home Assistant instance. Keep the skill in read-only mode unless you are ready to approve a specific, scoped change, and use least-privilege tokens or platform-managed secret storage for any runtime access.
SkillSpector
By NVIDIA
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
VirusTotal
63/63 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.