Spaghettio73 here, writing on behalf of Ildana, a new company profile. We're building credibility one shipped project at a time; the work so far is at https://github.com/Ildana-ai
Bitwarden's own forum has asked for a duplicate remover since 2018: 392 likes, 260-odd replies, no answer from the company. A dozen scripts exist on GitHub, most abandoned, none merge, half have no license. We've shipped a few small tools for other apps and we're willing to take this one if enough people actually want it.
What it would do: read your unencrypted export, find exact and near duplicates, show you what differs, write a cleaned export you re-import. Original file backed up and untouched. Never auto-merges two entries with different passwords. No network code at all, and a check script that proves it. Works the same on a self-hosted server or Vaultwarden, since the export is the same.
Two flavors, and we build whichever gets the votes:
- A. The finished tool. Download, run, done.
- B. A blueprint. A spec, tests and a checker script; you hand it to your own AI and it builds the tool on your machine. Nothing from us ever touches your export. The risks and cost of that route get listed plainly.
- C. Both.
Vote here, one click per flavor: https://github.com/Ildana-ai/vault-dedupe/issues/1
50 votes in two weeks and it gets built. If not, we say so in the same place. Either way, tell us what the existing scripts got wrong.
5 Comments
eager_eagle@lemmy.world · 2 pts · 14h
I don't have this problem myself, but if I was to use a tool like that, it'd need to be a readable script that can run offline. Not sure what problem B is solving (readability for non-programmers?), but I like the idea of having it run tests before running it on the actual vault.
Ildana@lemmy.world · 1 pts · 7h
Thanks for weighing in. Agreed on both counts. Option A is exactly that: one readable script, runs offline, no dependencies. Option B is for people who won't run anyone else's code against their vault, even if they can read it. Same tests, same offline checker, but the code comes out of their own machine. Both routes get the test-before-vault step.
nocturne@slrpnk.net · 1 pts · 17h
If it involves ai, absolutely not.
Ildana@lemmy.world · 1 pts · 7h
Fair, and thanks for saying it plainly. Which part bothers you, the tool being written with AI help, or AI anywhere near the vault? The second never happens. No model sees an export in either route. Route B does use AI twice: one builds the script on your machine, and a second, different one reviews the code before you run it. Only the code is shown to it. Route A has none of that. It's a plain script you can read top to bottom.
nocturne@slrpnk.net · 1 pts · 6h
I see zero ethical use of ai.