create-vo-elevenlabs

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime path scripts/gen_vo.py -> media_proxy.eleven_tts(..., text, voice_id, ...) sends caller/template-supplied --text into the ElevenLabs proxy request body as free text, so an outsider who can trigger the workflow can inject arbitrary text.