OpenClaw

openclaw-reflect

Self-improvement layer with evaluation separation, rollback, and tiered operator gates. Observes outcomes across sessions, detects recurring patterns, proposes improvements, validates proposals through a separate evaluator invocation, and applies changes safely with snapshot/rollback capability.

2.8k stars
openclaw/skillsskills/atlaspa/openclaw-reflectMarch 14, 2026
View on GitHub

Install command

python "$CODEX_HOME/skills/.system/skill-installer/scripts/install-skill-from-github.py" --repo openclaw/skills --path skills/atlaspa/openclaw-reflect
Tell me the task — I'll narrow the agent shortlist.