Smart Home IoT Control

• Outcome: Review
• Latest audit: Jun 13, 2026 at 1:13 PM UTC
• Version:
  1.0.0

Overview

The skill is coherent and cleanly packaged, but it gives agents broad smart-home control over locks, security systems, and appliances without clear confirmation or safety limits.

Review this skill carefully before installing. Use it only with accounts and devices you are willing to let an agent control, and add your own confirmation requirements for locks, alarms, thermostats, ovens, coffee makers, and other safety-sensitive appliances. Prefer limited-role credentials, audit logs, and narrow automation scopes.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill enables safety-critical actions such as locking doors, arming security systems, and controlling appliances via natural language, but it provides no warnings about unintended execution, device-specific hazards, or the need for confirmation on high-risk commands. In a smart-home context, ambiguous or mistaken commands can create physical security, safety, or property risks, making the omission materially dangerous.

VirusTotal

64/64 vendors flagged this skill as clean.