azure-search-documents-ts

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFE
Full Analysis
  • [Data Exposure & Exfiltration] (SAFE): The skill correctly manages sensitive information by using environment variables for the search endpoint and admin key. The documentation specifically lists 'Hardcoded credentials' as an anti-pattern to avoid and promotes the use of DefaultAzureCredential for secure authentication.
  • [Unverifiable Dependencies] (SAFE): All identified dependencies are official, high-reputation packages from Microsoft (@azure/search-documents, @azure/identity) installed via standard package managers.
  • [Indirect Prompt Injection] (LOW): As a search skill, it naturally processes user input for queries. However, it demonstrates the use of the odata helper (e.g., odatarating ge ${minRating}``) which provides proper sanitization/escaping for OData filter expressions, mitigating injection risks in the query surface.
  • [Command Execution] (SAFE): There are no instances of arbitrary command execution, shell spawning, or script evaluation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 04:29 PM