Hello all,
I am a french software engineer and I have been transitioning all of my accounting to bean count (it's great software, thanks !). I had some issues understanding how I should implement some stuff because I have some difficulty understanding english accouting terms.
I use nixos, so I am somewhat used to struggling with docs.
I have to say first that I am somewhat agnostic on AI for software dev, but one of my colleague said: "try it for docs it will be great". So I got the antitgravity free month on their pro plan and I tried to build a diataxis doc for beancount using opencode (as I understand it it's a clone of claude code).
Here is the result:
https://slashformotion.github.io/beancount-docs/The repo is here:
https://github.com/slashformotion/beancount-docsThe result is ok at best, I think some of the bql part of the doc is incorrect. But I completed this under an hour, which is interesting.
I used the docs and fava repo of the beancount org to feed the model, If you want me to take the site/repo down for IP I can no issues.
Curious to get your thoughts on the repo/site ? Do you think there is a point in having a docs beside the google docs (especially when it comes to syncing) ?
Cheers,