omnivoice

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow ingests user-provided free text via the local-first OpenAI-compatible TTS API parameter input to /v1/audio/speech (and user-provided audio for STT to /v1/audio/transcriptions), meaning outsider-authored text can be provided at runtime to the LLM/system.