ClawHub
Back to skill

Security audit

Alibaba Cloud AI Audio TTS Voice Clone

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Alibaba Cloud voice-cloning helper skill with sensitive but expected handling of voice samples, credentials, and local output files.

Install only if you intend to use Alibaba Cloud voice cloning and have clear consent for the voice sample. Use a dedicated API key, keep generated request files, audio URLs, and voice IDs private, and delete local outputs when they are no longer needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill documents file read/write behavior through helper scripts and output generation, but it does not declare corresponding permissions. This creates a transparency and governance gap: downstream systems or reviewers may underestimate the skill's ability to access local files and write artifacts, which can enable unintended data exposure or filesystem modification when the skill is executed.

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The manifest's default_prompt references a different skill identifier than the provided skill context, creating a mismatch between what the user selected and what the agent may actually invoke. This can cause unintended capability routing, confuse auditability, and potentially trigger the wrong external integration or credential scope if similarly named skills exist.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
This skill handles highly sensitive biometric-like voice samples by sending them to a remote cloud service and may return remotely hosted audio URLs, but it does not clearly warn users about data transmission, retention, or privacy implications. In the context of voice cloning, that omission is more dangerous than usual because users may unknowingly upload another person's voice or regulated personal data without informed consent or appropriate handling safeguards.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.