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). The SKILL.md explicitly uses client.analyze_from_url with arbitrary image_url inputs (see the "Analyze Image from URL" and other examples) so the agent fetches images from public URLs and processes OCR/captions/objects—untrusted third‑party content that could embed instructions and influence downstream behavior.