bframe is a wrapper around a duckdb connection which makes it easy to extend and understand.

The largest difference between bframe and alternatives is how we treat persistence. The library connects to an existing store (bring your own infra), while other solutions use their own storage.

This has several benefits if you are willing to provide your own store: a single source of truth (no weird syncing), the library can be directly used without a server (e.g. pip install bframelib), sources are generally fungible (use CSVs, JSON, postgres, etc).

bframe does a lot more which can be seen in our documentation: https://bframe.work

The github repository can be found here: https://github.com/bframe-work/bframelib

The project is still very much in development so I appreciate the feedback. If this project interests you and you'd like to chat more send me a message @ allan at bframe.work