ClawHub
Back to skill

Security audit

竖版视频生成免费版

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a video-generation helper, but it exposes broad custom command execution for TTS while also claiming commands are whitelisted.

Only install this if you trust the workflow and will review any `--tts-command` value before execution. Avoid using untrusted scripts or speech text with custom command templates, and prefer fixed, well-known TTS commands with carefully quoted arguments.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Intent-Code Divergence

High
Confidence
98% confidence
Finding
The skill's safety section claims only whitelisted commands are executed, but elsewhere it explicitly exposes `--tts-command` as an arbitrary command template. In a skill with `exec` permission, that contradiction can mislead users and agents into executing attacker-controlled shell commands through templated input such as text, output path, voice, or rate.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
Advertising a fully custom `--tts-command` makes the skill capable of arbitrary command execution beyond normal markdown-to-video generation. Because the skill is designed for agent use and has `exec` enabled, this expands the attack surface substantially: untrusted script content or user parameters may be routed into shell execution paths.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.