If you meant StemJSON as a prompt, it's not. Prompts are still free-form text, but the output is StemJSON - JSON created according to a specification that can describe complex logic along with the UI, and it can be rendered natively on iOS and Android. So a StemJSON module can be delivered as an SDUI payload, bundled as part of the app's functionality, or generated directly in the mobile app if it has access to AI.

No, I meant plain text as a description of a UI+logic. Why do you need a special syntax for that, when the audience is an LLM that understands text?

I guess there's a bit of a misunderstanding. The audience isn't an LLM. The purpose of StemJSON is to get a working native feature/module/app from just a prompt. An LLM can produce code, but that code can't be delivered directly to a published mobile app to extend it. Instead, the LLM produces output in a structured format describing the UI and behavior. The runtime in the app then deserializes it and builds native UI and logic from it, and that output can be reused over and over as standalone functionality.