azure-ai

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

• Third-party content exposure detected (high risk: 0.80). This skill exposes the agent to arbitrary external content because multiple SDK patterns accept and fetch external URLs/blobs (e.g., analyze_from_url(image_url=...) in references/sdk/azure-ai-vision-imageanalysis-py.md, Image from URL setBlobUrl(url) in references/sdk/azure-ai-contentsafety-java.md, batch transcription content_urls in references/sdk/azure-ai-transcription-py.md, and AnalyzeDocumentAsync(..., uri) in references/sdk/azure-ai-document-intelligence-dotnet.md), meaning untrusted third-party/user-provided content can be ingested and interpreted as part of the workflow.