azure-ai-voicelive-ts
Pass
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: SAFE
Full Analysis
- [Official SDK Integration]: The skill incorporates official Azure SDK packages such as @azure/ai-voicelive and @azure/identity, which are maintained for security and stability.
- [Secure Credential Management]: Guidance provided correctly steers users toward Managed Identity (DefaultAzureCredential) and environment variables, effectively mitigating the risk of hardcoded secret exposure.
- [Data Ingestion Analysis]: This skill processes external audio and text data (Ingestion points: session.sendAudio, session.addConversationItem; Boundary markers: SDK protocol delimiters; Capability inventory: Real-time speech-to-speech and function calling via Azure; Sanitization: Handled by Azure service infrastructure). While this represents an ingestion surface for potential indirect prompt injection, it is managed within the secure context of Microsoft's cloud services.
- [Resource Verification]: All documentation links and package references trace back to official Microsoft and Azure repositories, conforming to trusted source standards.
Audit Metadata