azure-ai-voicelive-java
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill enables an agent to process real-time bidirectional audio and text, which serves as an ingestion point for potentially malicious instructions hidden in external data.\n
- Ingestion points: Audio input is received via the
session.sendInputAudio()method, and session instructions are set throughVoiceLiveSessionOptions.\n - Boundary markers: The provided code snippets do not include specific delimiters or warnings to ignore embedded commands within the audio stream.\n
- Capability inventory: The skill establishes WebSocket connections and performs network operations with Azure OpenAI services.\n
- Sanitization: No input validation or sanitization mechanisms are implemented in the demonstration code.\n- [EXTERNAL_DOWNLOADS]: The skill references the official
com.azure:azure-ai-voiceliveMaven dependency and provides reference links to Azure's official GitHub repository for source code and samples.
Audit Metadata