ClawHub
Back to skill

Security audit

free-mp4-pptzc

Security checks across malware telemetry and agentic risk

Overview

This skill appears built for making narrated presentation videos, but it can modify the host by installing browser tooling and can send narration text to an external TTS service without clear consent controls.

Install only if you are comfortable with dependency and browser downloads, local browser rendering, ffmpeg processing, and narration text being sent to Edge TTS. Use an isolated environment, avoid confidential scripts unless external TTS is approved, and choose output paths carefully because existing files may be overwritten.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
with open(tmp_js, 'w') as f:
        f.write(script)
    
    result = subprocess.run(["node", tmp_js], capture_output=True, text=True)
    
    # 清理临时文件
    os.remove(tmp_html)
Confidence
90% confidence
Finding
The script generates a JavaScript file embedding unescaped file paths into string literals, then executes it with Node. If output_path or the temporary HTML path contains quotes or crafted characters, the generated JS can break out of the string and execute attacker-controlled code on the host.

Context-Inappropriate Capability

Medium
Confidence
83% confidence
Finding
The script can install Python packages and browser binaries at runtime, which changes the host environment and introduces supply-chain exposure beyond simple video generation. In an agent/skill context this is more dangerous because users may invoke it expecting local media processing, not software installation and network retrieval.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly recommends Edge TTS and notes that it requires network connectivity, but it does not clearly disclose that user-provided narration text may be sent to a third-party cloud service for processing. In a presentation workflow, that text may contain internal business updates, project details, or training content, so the omission creates a real privacy and data-handling risk rather than a purely cosmetic documentation issue.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The script sends slide text to edge_tts, which relies on an external service, without prominently warning that script contents may leave the local machine. If users process sensitive scripts, this can lead to unintended disclosure of proprietary or confidential content.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The command includes ffmpeg's -y option, which forces overwrite of the destination file without confirmation. If output_file is user-controlled or points to an important existing file, this can cause silent data loss or destructive overwrites in automated environments.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.