video1.0<iframe src="https://www.loom.com/embed/84c413168ba34c0ab794a54c9f552fc3" frameborder="0" width="1422" height="1066" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>10661422Loomhttps://www.loom.com10661422https://cdn.loom.com/sessions/thumbnails/84c413168ba34c0ab794a54c9f552fc3-21f1b4ef14f65ddc.gif88.213In this Loom I set out a plan for ChatGPT implementation and then the initial implementation, including the prompt builder from the context, the endpoint, service side events, and a basic skeleton. I tested it by adding an orange item and watching how streaming chunks build the context on top of what we provide. I also started another run and stopped the server to show errors. There is an issue with a double error indicator. No specific viewer action was requested.