Reverse engineered prompt

Build me a standalone version of this app, a simple voice first todo list where I can talk naturally and my tasks update live on the page. I want to be able to create lists, add tasks, edit them, reorder them, complete them, and delete them either by clicking in the UI or by speaking. When I turn on the mic, it should connect to a voice assistant, show a nice listening state with some audio activity feedback, and handle commands like “add milk to groceries” or “move meeting prep to the top.”

Please make it feel like a real usable demo from this repo, even though the README says some pieces were originally part of a larger app. If something is missing, fill in the gaps in the simplest way that keeps the core idea working. Changes should be saved and should instantly appear in every open browser window, not just the one I’m using. Keep the interface clean and minimal, with the voice experience being the main focus. Look up current docs online if you need to.

Want more depth? Deep Reverse