ClawHub
Back to skill

Security audit

YouTube ASR Summarize (Local)

Security checks across malware telemetry and agentic risk

Overview

This skill uses standard local media tools to summarize YouTube videos and writes outputs to a chosen folder, with no evidence of hidden exfiltration or destructive behavior.

Install only if you are comfortable running yt-dlp, ffmpeg, and faster-whisper locally and letting the tool contact YouTube to fetch metadata, subtitles, audio, and optional video frames. Use a dedicated output folder and delete generated transcripts or frames when they are no longer needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill advertises and documents shell execution, local file reads/writes, and environment use, but the manifest shown in SKILL.md does not declare corresponding permissions. This undermines transparency and sandbox/policy review because operators may approve the skill without realizing it downloads media, writes transcripts and frames to disk, and invokes external tools.

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding
The documented purpose says the skill is for videos with no subtitles and uses local ASR, but the behavior reportedly also fetches and prefers existing subtitles or auto-subtitles. This mismatch is security-relevant because users may choose the skill specifically to avoid certain data flows or assumptions, yet it may process external subtitle tracks and metadata in ways not disclosed.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill writes transcript text, summaries, and extracted video frames to local storage, but the description does not warn users about this persistence. Stored outputs can contain sensitive or copyrighted content and may remain in shared temp directories or backups, creating privacy and data-handling risks beyond the immediate task.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.