qwen3-tts

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill's docs and API reference explicitly allow passing external URLs as ref_audio or to the tokenizer (e.g., generate_voice_clone ref_audio can be a "Path, URL" and tokenizer.encode accepts "Path, URL") as shown in SKILL.md and references/api_reference.md, so the agent can fetch and ingest arbitrary public audio/web content (including GitHub) that could influence generation.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The fine-tuning workflow explicitly instructs cloning and running code from https://github.com/QwenLM/Qwen3-TTS.git (git clone ... then python ...), so remote code is fetched and executed as a required runtime dependency.