azure-ai-document-intelligence-ts

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches official Microsoft Azure SDK packages (@azure-rest/ai-document-intelligence and @azure/identity) to facilitate document analysis.
  • [DATA_EXFILTRATION]: Transmits document data to user-configured Azure endpoints for legitimate processing and extraction. This is standard behavior for the intended use case.
  • [PROMPT_INJECTION]: The skill acts as a processing surface for untrusted external data, which is a vector for indirect prompt injection (Category 8).
  • Ingestion points: Processes documents read from the local file system via node:fs/promises and remote documents via urlSource parameters.
  • Boundary markers: The provided examples do not implement delimiters or safety instructions to prevent the agent from obeying commands embedded within the documents.
  • Capability inventory: Includes local file read capabilities and network requests to external API services.
  • Sanitization: Extracted content is returned to the caller without sanitization or structural validation against malicious payloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 05:58 AM