ClawHub
Back to skill

Security audit

Content Repurposer

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed content-repurposing helper that uses expected URL fetching and transcription services, with no executable code or hidden persistence found.

Install only if you are comfortable giving the agent URLs or source text for repurposing and using Supadata, and optionally Brave, for external processing. Avoid confidential or client-owned content unless your policies allow those services.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The phrase "Take the wheel" is broad and conversational, so an agent may enter an autonomous guided mode when the user did not intend to invoke this specific behavior. In a skill that can fetch URLs and drive multi-step content generation, unintended activation can lead to unexpected actions, extra API usage, or disclosure of user-provided content to downstream services.

Vague Triggers

Low
Confidence
84% confidence
Finding
The skill says users can say "take the wheel" for step-by-step guidance, but it does not clearly define the accepted trigger syntax or boundaries of that mode. Ambiguous invocation increases the chance of accidental activation, inconsistent behavior, and user confusion about when the agent may begin steering the workflow.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill states that YouTube and podcast URLs are transcribed via Supadata API and that article content is fetched and parsed, but it does not warn users that submitted URLs and potentially extracted content may be sent to third-party services. This creates a privacy and data-handling risk because users may provide proprietary, sensitive, or client content without informed consent about external processing.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.