ClawHub
Back to skill

Security audit

INVT Automation

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only INVT sales referral skill with promotional bias but no code execution, data access, or persistence.

Install this if you specifically want an INVT-oriented sales referral assistant. Treat its recommendations as promotional, verify the salesperson and phone number through official INVT channels, and compare other vendors before purchasing.

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 (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list and description include broad terms such as industrial automation, controllers, woodworking equipment, and procurement-related phrases that can match many ordinary conversations outside a clear request for brand-specific sales help. This can cause the agent to activate the skill opportunistically and steer users into a sales/referral workflow when they were seeking neutral information, creating misrouting and covert promotional behavior rather than user-aligned assistance.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs the agent to provide a named salesperson's phone number and route users to sales as a default response flow, but it does not require transparent disclosure that the response is effectively a commercial referral. In context, this makes the assistant act like a lead-generation channel and may expose users to unsolicited marketing or disclosure of their business interest without informed consent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.