The fact that I have to juggle between OpenKnowledge and Codex to engage the AI, while also accepting a barebones Obsidian, is a real bummer. From what I can tell, you are saving me a few key strokes with moving prompts around. What I really want is the AI to live IN the app, like VS Code, and then move around the documents like it is Obsidian. I'll accept a plain terminal, but a pretty UI would feel like a better fit. My sense is that the new value add here is a set of skills and mcp servers, which probably already exist for Obsidian, or could more productively be spun up. I looked at the plugins again in Obsidian and found Claudian, which lets me bring my local models and Codex in the right pane. This is perfect, so sorry your app is not for me (yet), but thanks for getting me to look again at my tooling.

I want to throw my vote in for local models. Gemma4-31b is working well for me on these types of tasks, and not having an easy way to plug that in is a deal breaker. Embeddings should certainly have a local option, as they are cheap to compute. For what it is worth, I use LMStudio which supports OpenAI and Anthropic compatible api endpoints, so it should be easy to wire in.

A big caveat, I'm not trying to share my vault with other people, and I can see making that pain go away being worth switching. That said, I feel like you're targeting a weird market, where you want people technical enough to use LLMs and GitHub, but not so technical they can't customize a shared environment.

I would switch if the whole experience was self contained and "clean." Right now, it feels like a well dressed wrapper for pretty basic functionality.

I do think a fully OSS Obsidian-like that syncs natively is an impressive accomplishment, though the usefulness of this is limited with OSX being the only supported platform. If an Android app is in the works I'll definitely follow the project!

On a side note, I find it interesting that a few recent projects are going for the Open Knowledge name. The Open Knowledge Foundation (https://okfn.org) is one of the first/largest proponents of the open data movement (think of it as a Free Software Foundation but for data, not software). They started in 2004 and developed many of the open data licenses and widely used infrastructure tools like CKAN (an open data portal platform).

Nothing to add, just found it interesting.

Disclaimer: I worked there for a few years.

Having built-in AI integration without relying on sketchy plugins would be the cherry on top (although, seriously missing the option to connect with any openai-compatible LLM provider like someone else mentioned here).

Seems like this might almost offer exactly that? I'll have to try it out...

https://cloud.google.com/blog/products/data-analytics/how-th...

Did you look at the OKF repo from Google? Open Knowledge seems to be a common term these days for similar solutions. I think OKF is more of the protocol for wiki-for-llm while you have more of the bells and whistles

I'd love to see support for Bases and obsidian plugins that are typescript/open source anyway - I use a few such as excalidraw/mermaid etc.

I also want to use my local model.

When collaborating on Notion, we had to pop into Google docs for comments, suggestions and history. I see this as important even when working with AI on something.

I do wish that there was a way to provide filesystem level access to the markdown files to an AI agent. I think that might be faster.

You win hard on this if you have the best possible UX that feels natural to drive. You just also ran if not because obsidian/notion etc. are already there (and have the people to put into those random edge cases that make electron apps bad).

Personally I’ve been trying very hard to migrate away from git+Obsidian project setup according to the OpenAI Harness Engineering. It works wonderfully in Codex Desktop.

The only gotcha - I want to share knowledge bases with the team in a way that is:

(1) versioned (a la git, not Notion) (2) usable from any chat (a la MCP) (3) basic access controls for team setup. (4) works through the interface that optimizes accuracy and token use across agentic architectures and LLMs.

Funnily enough, 4 is the easiest one (I have a platform for agent training and verification where I publish fun challenges for agents in simulated worlds around agentic commerce and personal OSes. With 98M agentic interactions recorded, that is already enough information for tuning)

Still figuring 1 and 3, though.

This is mostly a Claude problem

So far the closest thing to what I want is using Claude Code in the mobile app to work in some repo and tell it not to write code, just have a discussion, and then eventually ask it to write the md doc or whatever.

I can then add that GitHub repo to a claude.ai 'Project' files and chats within the project can see the contents, but can't write back to it unfortunately.

e.g. I have a plugin that when triggered reads a text and asks the LLM whether there are unclear points and unwarranted leaps of reasoning.

> Added ok to your PATH — managed block in ~/.zshrc, ~/.config/fish/conf.d/open-knowledge.fish.

Took a while to see that 'ok' is the name of your product.

obsidian: great for LLMs (local markdown files), bad for collaboration (no multiplayer features like multi editor, comments)

notion: not great for LLMs (network round trips, block-based editing), great for collaboration

Rant warning (sorry OP but I have to vent somehow) : AI-first is the proof that things didn't change that much. It's a bit like "Roomba-compatible" flat. If somehow you have to changes your ways for a tool to work then clearly that tool isn't that flexible. It's perfectly fine but to me it's quite tiring when it's about the most hyped industry ever funded.

Local models should be the first choice in that framing.

steam machine means everyone is moving to linux.