ClawHub
Back to skill

Security audit

Lofy Home

Security checks across malware telemetry and agentic risk

Overview

This smart-home skill appears purpose-built and not malicious, but it can change real devices from broad natural-language phrases with limited upfront confirmation boundaries.

Install only if you are comfortable giving the agent control over your Home Assistant devices. Use a least-privilege Home Assistant token if possible, keep it on a trusted local network, review the quick-command phrases, and require explicit confirmation for HVAC changes, multi-device scenes, presence modes, and wake-on-LAN actions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The quick-command phrases are broad, conversational utterances such as "it's cold," "I'm home," and "goodnight" that can easily occur in normal dialogue without an explicit intent to control devices. In a home-control skill, this creates a real risk of unintended physical-world actions, including changing thermostat settings, activating scenes, or powering systems on via automation triggers.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill performs networked control of lights, media devices, thermostat behavior, and wake-on-LAN, but the documentation does not clearly foreground that these commands cause real system and environment changes. In this context, lack of user-facing warning and safety boundaries increases the chance of accidental or uninformed activation of actions with privacy, availability, and physical-world effects.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
1. Read `data/home-config.json` for device mappings and scenes
2. Confirm actions in ONE short message
3. If a device fails, report which one and suggest a fix
4. Never execute "turn off all devices" without confirmation
5. If Home Assistant is unreachable, report and suggest checking connection
6. Device entity_ids must be configured by user — prompt if missing
Confidence
88% confidence
Finding
The instructions only forbid confirmation bypass for one specific action, "turn off all devices," which implies many other impactful actions may be executed immediately and only confirmed after the fact. In a smart-home context, autonomous execution of scenes, presence-based actions, thermostat changes, media playback, and PC wake events can cause unintended device state changes without explicit user approval.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.