Interesting concept! But I'm struggling to understand what makes this strictly better than just having agents push to separate git branches and then having an orchestrator agent resolve the merge conflicts? I've been doing that for a few months with auto-PRs and it seems to work mostly fine, even if it's a bit slow. Is the lockless CRDT really necessary?
looks super cool, I have one very interesting point for you btw, You know how git tracks state, and i see that you do that too, but I feel what a "git for agents" should also do is track "Events" the why of things.
because ultimately reasoning gets lost. If i changed something, the state logs the change but why it got changed gets missed.
I know its a lot of overhead, but version managing reasoning is the next big thing.
I like to think of them as two clocks, state clock and event clock.