Also on that point about keeping humans in the loop on decisions, I've found following the Research-Plan-Implement process where we humans review at each of tho

Also on that point about keeping humans in the loop on decisions, I've found following the Research-Plan-Implement process where we humans review at each of those stages, to be really helpful. This doc describes the skill I use with my agents so they keep me looped in: https://gist.github.com/rjcorwin/296885590dc8a4ebc64e70879dc04a0f https://gist.github.com/rjcorwin/296885590dc8a4ebc64e70879dc... Then I use cook to iterate and explore during the AI led parts.