azure-ai-document-intelligence-dotnet
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes official Microsoft SDKs (
Azure.AI.DocumentIntelligenceandAzure.Identity) for document processing. These are legitimate, versioned packages from the standard NuGet registry. - [SAFE]: Authentication examples follow industry best practices, prioritizing
DefaultAzureCredential(Entra ID) over static API keys and correctly utilizing environment variables for configuration to prevent hardcoded secret exposure. - [INDIRECT_PROMPT_INJECTION]: The skill provides functionality to extract text and structured data from external documents (e.g., invoices, receipts, and IDs). This creates a standard ingestion surface where content within a processed document could attempt to influence an AI agent. The skill documents security considerations, such as verifying accuracy through confidence scores and handling sensitive personal information according to established business rules.
- [SAFE]: No obfuscation, persistence mechanisms, or unauthorized privilege escalation commands were detected in the instructions or provided code fragments.
Audit Metadata