azure-ai-vision-imageanalysis-py

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). SKILL.md includes calls to client.analyze_from_url(image_url=...) that fetch and analyze arbitrary public image URLs (including OCR/read and captions), so the agent would read and act on untrusted third-party content obtained from those URLs.