azure-ai-openai-dotnet
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFE
Full Analysis
- [Secure Authentication Patterns]: The skill correctly emphasizes the use of DefaultAzureCredential and Azure.Identity for production environments, which avoids the risks associated with long-lived API keys. It also explicitly warns against hardcoding credentials in the associated documentation.
- [Dependency Management]: Installation steps use standard NuGet commands for official Microsoft and OpenAI packages (Azure.AI.OpenAI, OpenAI), which is the established secure method for managing .NET libraries.
- [Data Processing Surface]: The skill includes patterns for handling user input and external data retrieval through Azure AI Search. While these are standard ingestion points for untrusted data in AI applications, the skill encourages the use of structured outputs and explicit citation handling to manage these inputs securely.
Audit Metadata