ClawHub
Back to skill

Security audit

Cheap

Security checks across malware telemetry and agentic risk

Overview

This is a text-only price-comparison skill with one disclosed travel-price example, and it does not install code, store data, or request privileged access.

Install if you want an agent to help compare visible prices. Avoid sharing unnecessary personal details, especially for the flight-ticket scenario, and treat any result as a current-price lead rather than a guaranteed final checkout price; confirm final price, fees, seller reliability, and terms on the platform yourself.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The README extends the skill from product price comparison into flight-fare search and ticket-buying advice, which is a scope expansion into a different domain with different data types, safety expectations, and third-party services. This can cause the agent to collect itinerary or travel-related user details and act outside its declared purpose, increasing privacy and misuse risk through undocumented capability creep.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The documentation does not warn users that their product or travel-related inputs may be used to query external Chinese platforms, which reduces transparency around where user data may be sent. While the file is only descriptive, the omission can lead to uninformed disclosure of shopping preferences or itinerary details to third-party services.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger phrases are broad and likely to match ordinary shopping language, which can cause unintended activation of the skill in contexts where the user did not explicitly request cross-platform scraping or comparison. This increases the risk of misrouting user requests, producing irrelevant or overconfident outputs, and applying this skill's assumptions to conversations outside its intended scope.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.