azure-ai-vision-imageanalysis-java

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md and examples (e.g., multiple analyzeFromUrl calls such as "Generate Caption from URL" and the Complete Application example using an imageUrl including a raw.githubusercontent.com URL) explicitly fetch and analyze images from arbitrary public URLs, causing the agent to ingest untrusted third-party content (including OCR'd text and captions) that could materially influence its outputs or actions.