I'm starting to see some more projects and research around self-improving agents, so thought I'd reshare this design pattern I introduced last year.

The "FOXY method" tracks output, generates a reflection, and feeds the most relevant ones when running future objectives, to help guide better tasklists.

Probably could be done well with graphs. Anyways, more on this in the quoted thread.

Source: x.com/yoheinakajima/status/1826376586305946102