RRSI: Regularized Recursive Self-Improvement of Agent Harnessesgithub.com10 points·simonpure··0 commentsOpen articleSaveView on HN