Back to skill

Security audit

Video Downloader

Security checks across malware telemetry and agentic risk

Overview

This skill mostly does what it says, but it can automatically install an unpinned Python package without explicit approval.

Install only if you are comfortable with a skill that runs yt-dlp and writes downloaded media locally. Prefer installing a pinned yt-dlp version yourself in a virtual environment before use, and review the URL, output path, file type, disk space, and legal rights for the content before running downloads.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (6)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
"""Install yt-dlp."""
    print("Installing yt-dlp...")
    try:
        subprocess.run(['pip', 'install', 'yt-dlp'], check=True)
        return True
    except:
        print("Error: Could not install yt-dlp")
Confidence
96% confidence
Finding
subprocess.run(['pip', 'install', 'yt-dlp'], check=True)

Lp3

Medium
Category
MCP Least Privilege
Confidence
96% confidence
Finding
The skill instructs use of a shell-executed Python script to download remote content, but the manifest does not declare corresponding permissions or safety boundaries. Undeclared shell capability reduces transparency and reviewability, making it easier for an agent to perform file writes and network-driven actions the user may not expect.

Context-Inappropriate Capability

Medium
Confidence
98% confidence
Finding
Allowing the skill to install yt-dlp via pip means the agent can modify the environment and fetch executable code from package sources at runtime. Even if intended as convenience, this exceeds a narrow downloader role and creates avoidable supply-chain and change-control risk.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The description is broad enough to match many generic requests about videos, downloads, audio extraction, or metadata saving, which can cause over-invocation. In an agent setting, this increases the chance the skill is triggered in contexts where the user did not clearly consent to downloading remote media or writing files locally.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill documentation explains how to download and save media but does not warn that it will write potentially large files to local storage and may extract audio or metadata from remote content. Without an explicit warning and confirmation step, users may not understand the storage, privacy, or operational consequences of invoking the skill.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The code installs a package after only printing a generic message, without obtaining explicit approval or describing the security consequences. In an agent setting, silent dependency installation can surprise operators and lead to unintended code execution and persistent environment modification.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.