ClawHub
Back to skill

Security audit

Video Cut

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed video-editing workflow that downloads or reads a user-supplied video, processes it locally, and writes edited media outputs.

Install only if you are comfortable with a skill that runs local media tools, downloads user-provided video URLs, and creates or overwrites files such as work/source.mp4, work/audio16k.wav, first_cut.mp4, and reports. Review output paths before running it on large or private media.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill clearly instructs the agent to use shell commands, read environment/tool availability, and read/write files, but it does not declare those permissions. That mismatch is dangerous because it hides the skill's true execution capabilities from the permission model and from users reviewing what the skill can do.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill performs network downloads, invokes shell tools like yt-dlp and ffmpeg, and writes multiple output files, yet it does not require an explicit user-facing warning or confirmation before these system-modifying actions. This can lead to unexpected network access, disk usage, and local file changes, especially when operating on user-supplied URLs or media paths.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
scripts/transcribe.py:12