📄️ Using XML Chat prompts
📄️ Chat Streaming
📄️ Chat with Vision
This sample demonstrates how to build chat-based applications with vision capabilities using Semantic Kernel for TypeScript together with OpenAI’s multimodal models. With just a few lines of code, you can enable your agents to see and reason about images, opening the door to powerful new scenarios such as image understanding, content analysis, or interactive assistants that combine text and vision.