Security audit
Security checks for vulnerabilities and agentic risk
This skill is a clearly scoped helper for preparing Alibaba Cloud CosyVoice voice-cloning requests, with privacy and consent considerations but no hidden or destructive behavior found.
Install only if you intend to use Alibaba Cloud Model Studio for voice cloning. Use audio only when you have the rights and consent to create a cloned voice, treat generated voice_id values as sensitive persistent resources, and check Alibaba Cloud retention and deletion controls before enrolling real voices.
The skill documents local helper and validation commands that read skill files and write artifacts under an output directory, but it declares no explicit tool scope such as permissions or allowed-tools. In an agent environment, missing tool restrictions can lead to broader-than-intended file access or writes if the runtime infers or grants capabilities implicitly.
The skill explicitly supports creating cloned voices and reusing the returned voice_id in later synthesis calls, which introduces session/resource persistence beyond a single interaction. Persistent voice identifiers can enable unintended reuse, impersonation, or long-lived handling of sensitive biometric-derived resources if not clearly governed.
# Model Studio CosyVoice Voice Clone
Use the CosyVoice voice enrollment API to create cloned voices from public reference audio.
## Critical model names
The skill description and guidance do not prominently warn that reference audio is uploaded to an external Alibaba Cloud API and that enrollment creates a persistent custom voice resource. Users may unknowingly submit personal or biometric voice data and create retained assets, creating privacy, consent, and compliance risks.
The script sets --language-hint to zh by default, which imposes a specific language/locale behavior unless the user explicitly overrides it. This is a natural-language policy concern because the file does not offer a neutral default or explain why Chinese is required for this skill.
No suspicious patterns detected.