azure-ai-transcription-py

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly ingests audio from external URLs/blob storage (see content_urls in the "Transcription (Batch)" example) and accepts streamed audio input (begin_stream_transcription / send_audio_file), so untrusted, user-provided audio could be transcribed and indirectly inject instructions that influence agent behavior.