Back to skill

Security audit

Publish Video Vocal Remover

Security checks for vulnerabilities and agentic risk

Overview

This skill gives straightforward ffmpeg instructions to reduce voices in video audio, with no hidden code or persistence, though users should avoid overwriting original videos without a backup.

Install only if you want a simple ffmpeg-based workflow for reducing speech frequencies in videos. Review the generated '_novoice' file before replacing anything, keep backups of original media, and only run the sudo package-install command if you trust your system package manager and actually need ffmpeg installed.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger description is overly broad and can activate on many loosely related requests, which increases the chance the agent invokes this skill in the wrong context. Because the skill alters media content and is framed around multiple paraphrases, weak activation boundaries can cause unintended processing or bypass more appropriate, narrower tools.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The README centers Chinese natural-language trigger phrasing without an explicit user language preference or opt-in, which can bias skill routing and response behavior in multilingual environments. This is mainly a safety and UX boundary issue: the agent may infer language-specific behavior from the skill description rather than the user's current context.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 20)May include surrounding context.

bash
# Ubuntu/Debian
sudo apt-get install -y ffmpeg

# macOS
brew install ffmpeg

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill instructs users to replace the original file with mv "INPUT_novoice.mp4" "INPUT.mp4", which can irreversibly overwrite the source media if the user follows it without preserving a backup. In this context, the command is not inherently malicious, but the lack of explicit backup, verification, and confirmation guidance creates a real risk of accidental data loss, especially because the skill is aimed at media editing where source files may be valuable.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.