ClawHub
Back to skill

Security audit

Wyze Vacuum

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Wyze robot-vacuum helper that can read status and send start, pause, or dock commands when the user asks.

Install this only if you want Codex/OpenClaw to be able to control your Wyze robot vacuum using your Wyze API credentials. Give explicit commands such as start, pause, or dock, and name the vacuum if you have more than one. Review where WYZE_ENV and WYZE_TOKEN_DIR point because they store or reference Wyze authentication material.

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
Findings (1)

Missing User Warnings

Medium
Confidence
82% confidence
Finding
The script can issue real-world vacuum control actions (start, pause, dock) immediately based solely on command-line input, with no confirmation, dry-run mode, or safety interlock. In an agent/automation context, this increases the risk of unintended actuation from mis-parsed prompts, ambiguous device selection, or unauthorized invocation, potentially causing nuisance, disruption, or unsafe operation around people, pets, or obstacles.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.