azure-ai-voicelive-dotnet
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the installation of packages from NuGet, a well-known and standard package registry. It references the official
Azure.AI.VoiceLive,Azure.Identity, andNAudiopackages. - [EXTERNAL_DOWNLOADS]: Provides reference links to Microsoft's official documentation (learn.microsoft.com) and the trusted 'Azure' organization GitHub repository.
- [CREDENTIALS_UNSAFE]: Authentication examples use
DefaultAzureCredentialand environment variables, adhering to cloud security best practices. No hardcoded secrets are present; placeholder tags like<your-api-key>are used correctly for demonstration purposes. - [INDIRECT_PROMPT_INJECTION]: As an AI-integrated tool, it possesses a standard attack surface for indirect prompt injection.
- Ingestion points: Data enters the agent context through
session.GetUpdatesAsync()which processes server-sent events (text and audio deltas). - Boundary markers: None present in the code examples.
- Capability inventory: Performs network operations via WebSockets and audio processing via the NAudio library.
- Sanitization: No explicit sanitization of model outputs is shown in the provided examples.
Audit Metadata