I currently extract bookmarks from Firefox and Safari and store them inside a local database. Then a cronjob saves them to Wayback machine if a prior check revealed that they are currently not.. donating regularly for that. Mine just makes sure that the pages are not lost, but yours enables offline reading.

I'm uncertain what the best mechanism is, there are so many ways to solve it. From filtering to recrawling for new content to enabling more advanced features, there are so many possibilities.

You should checkout Shaarli[0] that has a thriving community.

[0] https://github.com/shaarli/Shaarli/

I had started working on something similar to this, but without the Google Drive component. I wanted something where I could right click and "snag" a file, link or document and have it saved to a server I controlled.

It's not complete, mostly because the frontend is a mess, but the backend is able to save files, pages and links (https://gitlab.com/thebird/snag). Used Rust backend, Svelte and JS for the extension (of course).

Any chance you have this open-sourced or described in more detail somewhere?

It's too hackish, system dependent, and not feature complete yet. I plan to run it as a flask app on the local network as a more intuitive way of tagging and managing bookmarks... lot more to do.

one day this will be as famous as youtube-dl

This is great, thank you!

Your hacked together workflow sounds extremely close to how ArchiveBox works! You should give it a try ;)

Been there done that. Real issue is ranking the search results by relevance to the search query.