ClawHub
Back to skill

Security audit

Ai Intelligent Ai Model Marketplace

Security checks across malware telemetry and agentic risk

Overview

This skill is not clearly malicious, but it advertises model trading, deployment, and automatic settlement while sending users to run an external app without explaining safeguards.

Review the external repository and dependencies before installing. Run it only in an isolated environment first, and do not connect payment, marketplace, cloud, or deployment credentials unless the code, permission model, transaction controls, and rollback behavior are clear and acceptable.

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)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The skill description is broad and only advertises high-impact capabilities like publishing models, trading, evaluation, and one-click deployment without defining when the skill should activate or what user intent is required. In an agent setting, vague scope can cause the skill to be invoked for loosely related requests and may lead to unintended financial, operational, or deployment actions.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill explicitly references trading and one-click deployment, which are financially and operationally sensitive actions, but provides no warnings, safeguards, approval steps, or limits. If an agent uses this skill autonomously, a user could be exposed to unauthorized transactions, deployment of untrusted models, cost-incurring actions, or system changes without informed consent.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.