This project originally came from a trip to Korea where I learned about personal color analysis. The in-person sessions are interesting but often cost a few hundred dollars, so I wrote down the idea years ago of trying to approximate it with software.
I finally built a quick version as a learning project.
There’s nothing particularly sophisticated happening under the hood. The goal was mostly to get more comfortable working with modern AI tooling and to practice shipping small things again after a long stretch working inside a startup.
If anyone experiments with it, I’d be especially interested in:
• cases where the color analysis is clearly wrong
• UX things that feel confusing or clunky
• obvious technical improvements
• or just general thoughts on whether the idea is interesting at all
If the agent builds entity pages incrementally while reading, how do you prevent early incorrect assumptions about relationships or attributes from propagating through the entity graph? Is there support for belief revision?