azure-search-documents-dotnet

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or persistence mechanisms were detected. The skill provides legitimate code examples for using the official Azure AI Search SDK.
  • [EXTERNAL_DOWNLOADS]: The skill provides commands to install official Microsoft packages (Azure.Search.Documents and Azure.Identity) from the NuGet registry.
  • [DATA_EXFILTRATION]: Network operations target legitimate Azure service endpoints (search.windows.net and openai.azure.com) for searching and vectorization tasks.
  • [CREDENTIALS_UNSAFE]: Secure authentication practices are promoted, emphasizing the use of environment variables and Azure's identity framework over static API keys.
  • [PROMPT_INJECTION]:
  • Ingestion points: The skill ingests data from Azure Search indexes through searchClient.SearchAsync as documented in SKILL.md and references/semantic-search.md.
  • Boundary markers: The provided code snippets do not explicitly include delimiters to isolate ingested data from agent instructions.
  • Capability inventory: The skill allows for document management (UploadDocumentsAsync, DeleteDocumentsAsync) and index management (CreateOrUpdateIndexAsync).
  • Sanitization: Standard SDK usage is demonstrated without specific sanitization of retrieved search results.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 12:35 AM