ClawHub

video-clip-skill

Security checks across malware telemetry and agentic risk

Overview

This is a coherent YouTube clipping and subtitle skill, with a disclosed optional Groq transcription fallback that users should treat as a privacy-sensitive choice.

Install if you are comfortable with the agent running yt-dlp, ffmpeg, and Python to download/process YouTube media locally. Leave GROQ_API_KEY unset, or decline the Whisper fallback, when the audio is private, sensitive, copyrighted, or should not be sent to a third-party transcription API; only provide YouTube cookies when necessary and preferably through a temporary limited cookie file.

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 (2)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The trigger description is broad enough to match common requests about clipping, highlights, subtitles, or translation, which can cause this skill to activate in situations the user did not intend. In an agent ecosystem, overbroad activation increases the chance of unnecessary file operations, network calls to YouTube/Groq, and processing of user-provided media without sufficiently explicit consent.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill documents uploading audio to Groq for transcription but does not require an explicit user-facing privacy notice or affirmative consent at the moment data leaves the device. That creates a real privacy risk because spoken content from videos may contain personal, copyrighted, or sensitive information that users may assume stays local based on the skill's 'local' framing.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal