We’re live on Product Hunt.

Commit

3a5952d

My biggest issue today is that after an agent cooks, the review agent gives feedback to tune it. I often end up with a huge, over-engineered mess. So have the cut out the fat.
It depends; I spawn a few brand-new agents. I tell them Review the uncommitted work and write findings of bugs or issues into debug.md, then I include a relevant skill to give it focus PHP/LiveWire/Tests/Alpine, etc.. Once they are done, I hand it back to the main agent to decide.

commit 3a5952d