5 Comments
User's avatar
Yaroslav Ponomarev's avatar

Nice article! How did you implement the self-improvement into the agent? Is it fully automatic? And how does it identifies what and how to improve?

Bart Gasiorowski's avatar

No I pięknie :( już myślałem, że w weekend odpocznę, a teraz wiem co będę robił z moim _echo ;P

Didier's avatar

Very cool stuff!

At first it looks like Vibe-kanban or Auto Claude, but I guess it's closer to OpenClaw.

I'm very curious how you handle security / isolation ? Personally I'm only comfortable running all of that inside a VM, with a shared folder mounted on the host.

Pawel Jozefiak's avatar

It is closer to OpenClaw but it is not. Different architecture and tools. Fun to build!