azure-ai-vision-imageanalysis-py

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches official client libraries for Microsoft Azure services.
  • Installs azure-ai-vision-imageanalysis and azure-identity via pip.
  • [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection by processing external image data.
  • Ingestion points: SKILL.md demonstrates analyzing images from URLs (analyze_from_url) and local files (analyze).
  • Boundary markers: No delimiters or instructions are provided to the agent to treat OCR text or generated captions as untrusted data.
  • Capability inventory: No dangerous capabilities (like shell execution) are directly attached to this skill's output in the provided code, but the processed text becomes part of the agent's context.
  • Sanitization: No sanitization of the OCR text or caption results is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 05:17 AM