ClawHub
Back to skill

Security audit

Moltbot Home Assistant

Security checks across malware telemetry and agentic risk

Overview

This skill openly provides Home Assistant control for an agent, with meaningful safety controls, but users should configure it carefully because smart-home actions can affect the physical world.

Install only if you are comfortable giving an agent Home Assistant control. Use a dedicated Home Assistant token if possible, prefer HA_TOKEN over storing the token in config, keep safety enabled, add sensitive entities to blocked_entities, use allowed_entities to narrow access, and consider safety level 2 or broader critical_domains for scripts, automations, switches, climate, appliances, and scenes that may have safety or security impact.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Missing User Warnings

Low
Confidence
76% confidence
Finding
The README demonstrates write-capable home automation actions such as turning devices on/off, invoking scenes, and calling arbitrary Home Assistant services, but it does not prominently warn that these can cause physical-world effects beyond the limited 'critical domains' list. In this skill context, users or agents may infer that anything outside locks/alarms/covers is low risk, even though climate, scripts, switches, and scenes can also trigger impactful automations or unsafe device behavior.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
| Level | Behavior |
|-------|----------|
| 0 | No safety checks (⚠️ dangerous!) |
| 1 | Log actions only, no confirmations |
| 2 | Confirm all write operations |
| 3 | Confirm critical operations only (🏅 **recommended**) |
Confidence
88% confidence
Finding
no confirmation

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.