azure-ai-projects-dotnet

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides installation instructions for official Microsoft Azure SDK packages (Azure.AI.Projects, Azure.Identity, Azure.AI.Projects.OpenAI) from the NuGet registry, which is a well-known and trusted service.
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection through its data ingestion and agent execution capabilities.
  • Ingestion points: The Datasets section in SKILL.md provides methods to upload local files and folders (UploadFile, UploadFolder) into the Azure AI environment.
  • Boundary markers: The provided C# code snippets do not include explicit boundary markers or system instructions to the AI to ignore embedded commands within ingested datasets.
  • Capability inventory: The skill provides access to create and run persistent and versioned agents (CreateAgentAsync, CreateRunAsync, CreateResponse), which represents a mechanism for the AI to potentially follow instructions present in the ingested data.
  • Sanitization: The code snippets do not demonstrate sanitization or validation of the contents of files before they are uploaded and processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 05:34 PM