https://github.com/EL-MTN/LMUI-Protocol

Much like how servers used to render HTML, LMUI is a protocol that guides LLM apps/clients to render specified components. I'm not sure if anyone's already built an existing protocol, but I really think pure text is killing the potential of LLM clients, especially with how much more interactivity MCP's have brought for us to play with existing services. Improving the UI/UX should be a next step in LLM applications. PR/feedback is very welcome.