菜单
Read files surgically — grep first, read only relevant sections. Prevents loading entire large files when only a few lines are needed. Activates automatically when smart-read mode is on. Input token savings: 40–90% depending on file size and task scope.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Creates a structured session checkpoint that captures all critical context in minimal tokens. Use /checkpoint to save state before starting a new sub-task, switching files, or when context window is filling up. Typical compression: replaces 8,000–40,000 input tokens with ~400–800 tokens.
Analyzes current conversation context and identifies what can be safely dropped. Triggers a compressed summary that replaces bloated history. Invoke with /trim. Average input token savings: 55–70% on long sessions.
Output mode that shows only changed lines as diffs instead of rewriting full files. Cuts output tokens 60–80% on code edits. Works with Claude Code, Codex, Cursor, Windsurf. Activate with /diff-only. Deactivate with /diff-off or "show full file".
| name | smart-read |
| description | Read files surgically — grep first, read only relevant sections. Prevents loading entire large files when only a few lines are needed. Activates automatically when smart-read mode is on. Input token savings: 40–90% depending on file size and task scope. |
Never read an entire file unless the task genuinely requires full-file understanding.
Before reading any file, follow this order:
Grep first — search for the specific symbol, function, or pattern needed:
grep -n "functionName\|ClassName\|keyword" path/to/file
Read targeted range — use line numbers from grep to read only the relevant section:
sed -n '45,80p' path/to/file
Read ±10 lines around the target for context.
Full read only if:
| File size | Default action |
|---|---|
| < 50 lines | Read full |
| 50–200 lines | Grep first, read relevant sections |
| 200–1000 lines | Grep first, read targeted ranges only |
| > 1000 lines | Grep first, read max 100 lines at a time |
| File size | Normal read | Smart-read | Savings |
|---|---|---|---|
| 200 lines | ~1,400 tokens | ~280 tokens | ~80% |
| 500 lines | ~3,500 tokens | ~420 tokens | ~88% |
| 1000 lines | ~7,000 tokens | ~490 tokens | ~93% |
User says "read full file" or "smart-read off" → comply for that file, resume smart-read after.