analyze
CommandAnalyze user corrections and log them into the error notebook
Fully load memory files into the current conversation context.
Fully load memory files into the current conversation context.
SessionStart hooks inject a brief summary automatically, but `/reload` does the **full read** -- loading actual file contents so Claude can reference specific details.
1. **Read MEMORY.md** -- Scan all index entries (usually auto-loaded, but confirm) 2. **Read task file** -- Load `memory/todo-status.md` for current task status 3. **Read recent changes** -- Load files updated in the last 24 hours (up to 3 files) 4. **Report** -- "Loaded. {N} index entries, {M} pending tasks, recently changed: {summary}"
If the user specifies a topic:
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.