Would definitely try this out, but am Windows only.
Tried to create something like this but the best I got was a bookmarklet to copy the chat text. Ended up just exporting data and using that method for a one time pull.
I don't use AI coding tools very much, but the conversation aggregation/search feature sounds useful. If anything I'm interested in trying this just as a replacement for Anthropic's own search feature – their electron app is utterly buns, and I've been looking for a native macOS app to use in place of it.
Would definitely try this out, but am Windows only.
Tried to create something like this but the best I got was a bookmarklet to copy the chat text. Ended up just exporting data and using that method for a one time pull.
I'd also try it, but I'm Linux only ^_^
(thinking about cross-platform version, recalling my Java past)
Java (modern Java) would be awesome!
sorry about that :(
This is really cool, and also feels like it should be built into Claude Code (that’s a compliment btw). I’m gonna try it out this week, thanks.
Also Bonsa looks like the future congrats
Thank you :-)
Do you find any issues with FTS5 where it will miss some semantic detail?
Nope, at least not yet
I don't use AI coding tools very much, but the conversation aggregation/search feature sounds useful. If anything I'm interested in trying this just as a replacement for Anthropic's own search feature – their electron app is utterly buns, and I've been looking for a native macOS app to use in place of it.
Please try & let me know what you think - thanks
I dont get the point of local-first here? Like why would it not be local and why is it a feature?
Also isnt the data being pulled from the LLM firms' servers?
Like what is the meaning of this? You dont take all my chats and put it on your servers and steal my data? That is what local-first is these days?
Yes, thats what I meant. Sorry if it is misleading
Can you please clarify with your comment?
I believe i cant edit that comment