Autonomously improve a real artifact (code, training recipe, agent harness, data pipeline, prompt) against an objective and an evaluator, using Hypothesis Tree Refinement (HTR) from the Arbor paper. Use this whenever…

Agent skill published as a SKILL.md in K-Dense-AI/scientific-agent-skills (folder skills/arbor). Install it with one command using the open skills CLI, or copy the folder into ~/.claude/skills/.