azure-ai-contentsafety-py

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

• Third-party content exposure detected (high risk: 0.70). The skill's "Image from URL" example (AnalyzeImageOptions with ImageData(blob_url="https://example.com/image.jpg")) and the Analyze APIs show ingesting external/public content (image URLs and user-provided text) that the agent interprets to determine severities and actions like halt_on_blocklist_hit, so untrusted third‑party content can materially influence behavior.