azure-ai-contentsafety-ts

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

• Third-party content exposure detected (high risk: 0.80). The skill's Analyze Image example accepts arbitrary blob URLs (e.g., "https://storage.blob.core.windows.net/container/image.png" in the "Analyze Image
• Blob URL" section) and its /text:analyze endpoint processes arbitrary text input, so it fetches/ingests untrusted third-party content that the agent interprets and which can materially influence moderation decisions.