Deciduous: Better programming with LLMs using a living memory and decision graph (notactuallytreyanastasio.github.io)

Deciduous is built using itself. It tells the story of that here

http://notactuallytreyanastasio.github.io/deciduous/story.ht...

There is also a living graph of the entire development lifecycle thus far here

http://notactuallytreyanastasio.github.io/deciduous/demo/

I began the project 12 days ago, so this is all very fresh. I extracted it from another tool I built after building it to help me build that tool.

Looks very intriguing!

My main question after looking it over though is: given an existing codebase, how could I go about building a decision graph to essentially get "caught up" to current (I guess based on commit history?) to then move forward actively using Deciduous from here on out?