Captures learnings, errors, and corrections for continuous improvement. Use when: (1) A command fails unexpectedly, (2) User corrects you ('No, that's wrong....
Streamlined SKILL.md (231 lines, was ~800). Fixed clawdhub→clawhub command. Added trigger words to description. Split verbose multi-agent content into references/. Added skill-extraction and other-agents reference files. Added OpenClaw inter-session tool examples. Improved logging format with real examples.