azure-ai-contentsafety-java

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

• Third-party content exposure detected (high risk: 0.70). The skill's examples show the client ingesting arbitrary user/third-party content (e.g., AnalyzeImageOptions with ContentSafetyImageData.setBlobUrl("https://example.com/image.jpg") and AnalyzeTextOptions that accept arbitrary text), meaning the agent will fetch and read untrusted public URLs and user-provided text as part of its workflow.