ClawHub
Back to skill

Security audit

Voice Clone

Security checks across malware telemetry and agentic risk

Overview

The skill mostly does what it says, but its audio tool can run an unsafe shell command using a user-chosen output filename.

Review or patch the script before installing. Remove the automatic os.system xdg-open call or replace it with a non-shell subprocess call, and avoid unusual or untrusted --output filenames until fixed. Use OpenAI or ElevenLabs only for text and voice material you are comfortable sending to those services.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

os.system() or os exec-family call

High
Category
Dangerous Code Execution
Content
# 尝试播放 (如果可用)
        try:
            os.system(f"xdg-open '{output_file}' >/dev/null 2>&1 &")
        except:
            pass
Confidence
95% confidence
Finding
os.system(f"xdg-open '{output_file}' >/dev/null 2>&1 &")

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger phrases include very generic terms such as 'TTS', '朗读', and '生成语音', which can overlap with normal user requests and cause the skill to activate unexpectedly. In a skill that can send text to external TTS providers and perform voice-cloning operations, accidental invocation increases privacy and consent risks because users may not realize their content is being processed by third-party services.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill advertises integration with OpenAI, ElevenLabs, and other external TTS engines but does not warn users that submitted text, audio, or reference voice samples may be transmitted to third-party services. This omission can mislead users about data handling and consent, which is particularly sensitive in a voice-cloning context where personal or biometric-like voice data may be uploaded.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.