ClawHub
Back to skill

Security audit

Car Consult

Security checks across malware telemetry and agentic risk

Overview

This is a text-only electric-car buying advice skill with disclosed web search and narrow reference-file use, and no evidence of hidden execution, credential access, persistence, or data exfiltration.

Safe to install if you want a Chinese/Wuxi-oriented new energy vehicle buying advisor. Be aware it may activate for some broad car-cost or financing questions, and current price or policy claims should still be checked because the skill relies on web search and market data that can change.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill’s trigger scope is unusually broad and explicitly includes implicit cases, which increases the chance it will activate for loosely related vehicle, cost, or comparison queries that the user did not intend to route into this workflow. Over-broad activation can cause incorrect context capture, unnecessary file/tool access, and response shaping that overrides a more appropriate general assistant behavior, making this a genuine prompt-scope vulnerability even though it is not directly malicious.

Natural-Language Policy Violations

Medium
Confidence
81% confidence
Finding
The workflow hardcodes Chinese-language questioning and phrasing without checking the user’s preferred language, which can override user intent and reduce transparency about what the system is doing. While this is not a classic security exploit, it is a real safety and usability issue because forced language can mislead users, impair consent, and increase the likelihood of incorrect decisions in a purchase-advice context.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger set uses broad shopping and comparison language around cars without consistently anchoring activation to new energy vehicles only. This can cause the skill to activate on adjacent automotive requests, leading to misrouting, irrelevant advice, and possible suppression of better-matched skills, especially because many user queries omit exact drivetrain terminology.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.