Video Transcript Downloader

• Outcome: Review
• Latest audit: Jul 10, 2026 · 2:44 AM UTC
• Version: 1.0.0

Overview

The skill mostly does what it says, but it exposes an unrestricted yt-dlp argument passthrough that can be used to run powerful or unsafe yt-dlp behaviors.

Install only if you are comfortable with a skill that can run yt-dlp/ffmpeg and save remote media locally. Review every command before execution, choose output directories deliberately, and do not let untrusted page content or prompts supply extra yt-dlp arguments after --, especially options that execute commands or change output paths.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 83% confidence
• Finding: The skill advertises and invokes local command-line tooling (`yt-dlp`, `ffmpeg`, `npm ci`) and may rely on environment/system capabilities, yet it declares no permissions or warning boundaries. This creates a trust gap where an agent could use filesystem, network, or tool access without explicit user-visible permission framing, increasing the risk of unintended local actions.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger phrases are broad (`download this video`, `save this clip`, `get transcript`, troubleshooting yt-dlp/ffmpeg) and overlap with common user intents, which can cause the skill to activate in situations the user did not specifically intend. Because this skill downloads remote content and may execute external tools, accidental invocation raises the chance of unreviewed network access and local file creation.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The description does not clearly warn that the skill downloads untrusted remote content to local storage and may invoke external tools like `yt-dlp` and `ffmpeg`. In context, this is more dangerous because the skill’s primary purpose is retrieving remote media and writing files locally, so users and orchestrators need explicit notice before those side effects occur.

Unpinned Dependencies

Low

• Category: Supply Chain

Content

"private": true,
  "type": "module",
  "dependencies": {
    "youtube-transcript-plus": "^1.1.1"
  }
}

• Confidence: 94% confidence
• Finding: "youtube-transcript-plus": "^1.1.1"

VirusTotal

64/64 vendors flagged this skill as clean.

Static analysis

Detected: suspicious.dangerous_exec, suspicious.env_credential_access

Shell command execution detected (child_process).

Critical

• Code: suspicious.dangerous_exec
• Location: scripts/vtd.js:77

Environment variable access combined with network send.

Critical

• Code: suspicious.env_credential_access
• Location: scripts/vtd.js:61