I didn't set out to build a microkernel architecture.
I simply kept solving implementation problems, and many of the solutions looked surprisingly familiar in retrospect.
I'm especially interested in hearing from people who have built autonomous agents with different architectural approaches.
(The README and my comments were translated from Japanese with AI assistance.)