Taras Boiko

Self-improvement loop for Hermes Agent: reads its own trajectory, proposes and applies the smallest skill/memory edit (create/patch), journaled with rollback. /refine command, refine_run tool, on_session_end hook.

OpenClaw plugin that stops an agent repeating the same tool mistake: finds failures that recur across sessions, writes one short lesson, and puts it in front of the agent.