ClawHub
Back to skill

Security audit

Intiface Direct Control

Security checks across malware telemetry and agentic risk

Overview

The skill openly controls intimate hardware, but it can issue immediate physical actuation commands over local or LAN WebSocket without built-in confirmation, safety warnings, or value checks.

Install only if you intentionally want an agent to control Intiface-connected intimate devices. Keep Intiface bound to localhost unless remote access is truly needed, use only trusted networks, confirm the exact device before actuating it, start at low values, and ensure everyone affected has explicitly consented and can stop the device immediately.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The README provides immediate hardware-activation commands such as vibrate and stop without any safety warning, consent notice, or reminder that connected devices may activate physically as soon as commands are run. In the context of intimate-device control, this increases the risk of accidental or non-consensual activation, especially for users who copy-paste commands without understanding their effect.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The documentation instructs the agent to connect over WebSocket to control intimate devices, including over a LAN, but does not warn users about the privacy, consent, and physical-safety implications of network-based remote control. In this context, omission matters because misuse, unintended access, or control by the wrong host could cause non-consensual activation, exposure of sensitive device presence, or distress, making the skill more dangerous than ordinary IoT control guidance.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The script sends real-world device actuation commands immediately based on CLI arguments, with no confirmation prompt, consent check, safety interlock, or bounds validation before issuing vibration commands. In this skill context, that is more dangerous than a normal automation script because it directly controls intimate hardware, so accidental invocation, misuse by another local process, or socially engineered command execution can cause unwanted physical effects.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.