azure-ai-projects-py

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly exposes agents to open web content via tools and connections—see "Tools Overview" (BingGroundingTool: Web search) and the agent creation examples and "Prefer connections for external service integration (AI Search, Bing, etc.)" which show the agent ingesting and using third‑party web/search content that can influence its actions.