ClawHub
Back to skill

Security audit

car-recommender

Security checks across malware telemetry and agentic risk

Overview

This is a simple car-buying advice skill with no code, permissions, or sensitive access; its main issue is that broad trigger words may activate it unintentionally.

Safe to install as a conversational car-shopping helper. Be aware it may activate on broad Chinese terms, and verify current prices, incentives, sales rankings, and safety details with official sources or local dealers before making purchase decisions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger list includes very broad terms such as "车", "预算", and generic car-category words, which can cause the skill to activate in many unrelated conversations. This creates routing ambiguity and unintended invocation risk, especially in a shared assistant environment where user prompts may mention cars incidentally rather than explicitly requesting this skill.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.