IoT智能设备管理

• Outcome: Review
• Latest audit: Jun 12, 2026 at 7:22 AM UTC
• Version:
  1.0.0

Overview

The skill appears related to IoT assistance, but it asks for broad shell execution without enough clear boundaries for when commands may run.

Review this skill before installing. Use it only if you are comfortable granting shell access for IoT diagnostics, and require explicit confirmation before any command runs, especially commands involving local networks, device control, credentials, or configuration changes.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (2)

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: The skill requests shell execution capability even though the documented behavior only describes device management guidance, configuration generation, diagnostics, and reporting. Unnecessary shell access materially increases attack surface because a prompt-injected user request or malformed workflow could cause command execution on the host or connected environment.

Vague Triggers

Medium

• Confidence: 80% confidence
• Finding: The invocation description is broad enough to let the skill engage on many loosely related IoT requests without well-defined boundaries. In a skill that also declares powerful capabilities, vague triggering increases the chance of unintended activation, overreach into risky tasks, and exploitation through prompt manipulation.

VirusTotal

62/62 vendors flagged this skill as clean.