ClawHub
Back to skill

Security audit

Alicloud Ai Audio Cosyvoice Voice Clone

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Alibaba Cloud voice-cloning helper, but it needs user review because it creates reusable cloned voices from audio URLs without enough consent, privacy, or lifecycle guidance.

Install only if you will use it with audio from speakers who explicitly authorized cloning. Treat both the Alibaba Cloud API key and returned voice_id as sensitive, prefer time-limited sample URLs, avoid private or non-consensual recordings, and confirm provider retention/deletion options before creating long-lived voice clones.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill instructs users to submit public audio URLs and configure API credentials, but it does not clearly warn that reference audio will be transmitted to Alibaba Cloud and may contain biometric voice data or personal information. Because voice cloning involves sensitive identity-linked data, missing privacy and consent guidance increases the risk of unauthorized data sharing or misuse.

Session Persistence

Medium
Category
Rogue Agent
Content
# Model Studio CosyVoice Voice Clone

Use the CosyVoice voice enrollment API to create cloned voices from public reference audio.

## Critical model names
Confidence
89% confidence
Finding
create cloned voices from public reference audio. ## Critical model names Use `model="voice-enrollment"` and one of these `target_model` values: - `cosyvoice-v3.5-plus` - `cosyvoice-v3.5-flash` - `c

VirusTotal

53/53 vendors flagged this skill as clean.

View on VirusTotal