Why kbasis exists

Understanding is the bottleneck

AI keeps getting better. What it can't do for us is understand. Information arrives all day, and the hard part is no longer finding it. It is taking it in quickly, keeping it, finding it again, sharing it, and handing it to an AI that can act on it.

A structure you can search

The tools at hand were not built for that. Note apps manage documents, though most of what people keep now is written by an AI. Markdown lasts and agents write it natively, but it carries no meaning of its own. A graph of backlinks looks impressive and says little unless someone builds a data model on top of it, and almost nobody does. kbasis is a place where your AI can search, store and help you understand, on a graph whose links have types: a structure you can search without designing a data model yourself.

Who it's for

People who think alongside an AI every day and want to understand things faster. People who would rather build on a tool, make it their own and run it themselves than rent one.

Why it's open

If the best note app had been open source, there would already be a shared standard for how agents read and write knowledge. There isn't one yet, so kbasis is open, all of it, under the MIT licence: the code, the data model and the tools your AI calls. Take it, fork it, build on it. It runs on your own machine first. A hosted version may come later, second, and anyone else is free to host kbasis too.

What kbasis promises

  1. It stays MIT.Never relicensed to something more closed.
  2. Nothing is withheld.A hosted version never gets a capability the open one lacks.
  3. Your knowledge stays yours.Open, readable formats you can take anywhere.
  4. Any AI is your choice.Any client, any model, your own key.
  5. It's built in the open.Decisions and their reasons are public.
Read the full story