deepgram-dotnet-audio-intelligence

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

• Third-party content exposure detected (high risk: 0.75). The required runtime path is calling Deepgram’s REST/WebSocket transcription endpoints (e.g., TranscribeUrl(new UrlSource("https://..."))), where the LLM ingests the returned transcript/summary text that originates from the user-supplied external audio URL/content (outsider-authored free text via fetched media transcription results).