analyze
CommandAnalyze user corrections and log them into the error notebook
Check the health status of all memory files and hooks.
Check the health status of all memory files and hooks.
1. **Scan memory directory**: `~/.claude/projects/{project-id}/memory/` 2. **Scan sessions directory**: `~/.claude/sessions/` 3. **Scan diary directory**: `~/.claude/sessions/diary/` 4. **Scan learned directory**: `~/.claude/skills/learned/` 5. **Check debug log**: `~/.claude/sessions/debug.log`
Memory Health Report
MEMORY.md level: {lines}/200 ({percent}%) {Safe/Warning/Critical}
Memory files:
| File | Lines | Last Updated | Status |
|------|-------|-------------|--------|
| {filename} | {lines} | {date} | {OK/Stale/Too Large} |
Sessions: {N} total, latest: {date}
Diary entries: {N} total, latest: {date}
Pitfall records: {N} total, latest: {date}
Learned Skills: {N} total
Hooks status:
- session-start.js: {OK/Error}
- session-end.js: {OK/Error} (debug.log last: {msg})
- memory-sync.js: {OK}
- write-guard.js: {OK}
- pre-push-check.js: {OK}Claude Code 的記憶系統 | A memory system built with hooks + markdown. Zero dependencies.
Analyze user corrections and log them into the error notebook
Pause the SessionEnd auto-commit hook when refactoring core engine files.
Help decide when to use `/compact` based on the current conversation state.
Review the error notebook — flip through past corrections before working
Implement tasks incrementally — build, test, verify, commit. Add "auto" to run the whole plan in one approved pass.