
Lesson
This one captures insights from your conversation and stores them in two layers: a technical fact (the specific pitfall, cause, and fix) and a decision principle (the behavioral rule you should follow next time). It scans recent context when you trigger /lesson, writes both entries to memory with high importance scores, verifies they're retrievable, then reports back what got saved. The dual-layer approach is smart because you get both the tactical details and the strategic pattern, which means you're less likely to hit the same issue twice. If it detects the lesson should also update a checklist or skill file, it handles that too.
- skills.sh installs
- 335
- GitHub stars
- 4.4K
- last updated
- Jun 3, 2026
Install
$ npx -y skills add win4r/memory-lancedb-pro --skill lesson --agent claude-codeInstalls into .claude/skills of the current project.
Files
Lesson Extraction & Storage
When triggered, extract and store lessons from the recent conversation context.
Steps
- Scan recent context — identify the pitfall, bug fix, or key insight just discussed
- Store technical layer (category: fact, importance ≥ 0.8):
Pitfall: [symptom]. Cause: [root cause]. Fix: [solution]. Prevention: [how to avoid]. - Store principle layer (category: decision, importance ≥ 0.85):
Decision principle ([tag]): [behavioral rule]. Trigger: [when]. Action: [what to do]. - Verify —
memory_recallwith anchor keywords to confirm both entries retrievable - Report — tell Master what was stored (brief summary)
Rules
- Keep entries short and atomic (< 500 chars each)
- If the lesson also affects a checklist or SKILL.md, update those files too
- If no clear lesson is found in recent context, ask Master what to store
Sponsors
Details
- Repository
- github.com/win4r/memory-lancedb-pro
- Category
- AI & Agent Building
- First seen
- Jun 3, 2026