I love this idea! I agree that LLMs often don't have the knowledge that's actually important (a frustration I have with ChatGPT is that it remembers unimportant things and then over-indexes on them in its responses).
Totally agree. The current memory implementations feel a bit meh. you don’t really control what’s stored or how it’s used.
I’ve been thinking of it more as a visual knowledge layer you can inspect + operate on, rather than something implicit inside the model.
Curious if you’ve found any tools that handle this well today?
obsidian is overkill for most people. however, if you are managing more than private repo, you would like to see one place for your ai agents + data, a KB +LLMs.
Also, Cabinet is open source, it’s like Paperclip+obsidian together, open source and free. plus, as an engineer I feel very uncomfortable running closed software these days,as one skill can take over my computer.
yes, Cabinet core principle is BYOAI (Bring your own AI). Cabinet should work with Claude, Codex, OpenCode, local models, and whatever comes next, without forcing your knowledge into a single provider’s ecosystem.
Thanks! It's just 2 days old and i'm working full speed to merge all PRs. it's really exciting have my first open source project after years in corporates.
glad you liked it!
Yes, I’m thinking about org support, but trying to do it without losing the “files on disk / git-backed” model.
directions i'm thinking about: shared knowledge base (still file-based), permissions on top, optional hosted layer.
Still early though. curious if you’re using it solo or already trying to share it with a team?
I'm happy to offer more feedback- my email is micah.blachman@gmail.com