Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Claude Code 第二大腦三層結構設定。說「建立第二大腦」時載入。
.claude/skills/mathruffian-dot-cc-second-brain/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -14% | 0% |
建立三層結構：每日筆記/ → 創作庫/ → 知識庫/
每週重整：整理每日筆記 → 更新專案進度 → 清理無用筆記。
回報：各層級狀態、CLAUDE.md 是否已記錄。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 19,289 | 21,332 | +11% | 1 | 1 | 0% | 3,142 | 2,873 | -9% | 0 | 0 | — |
case-02 | fail→fail | 10,818 | 3,801 | -65% | 1 | 1 | 0% | 1,692 | 388 | -77% | 0 | 0 | — |
case-03 | fail→fail | 12,217 | 3,881 | -68% | 1 | 1 | 0% | 2,096 | 434 | -79% | 0 | 0 | — |
case-04 | fail→pass | 14,093 | 5,403 | -62% | 1 | 1 | 0% | 2,270 | 1,058 | -53% | 0 | 0 | — |
case-05 | fail→pass | 13,790 | 14,934 | +8% | 1 | 1 | 0% | 1,961 | 2,125 | +8% | 0 | 0 | — |
case-06 | pass→pass | 11,635 | 19,617 | +69% | 1 | 1 | 0% | 1,794 | 2,714 | +51% | 0 | 0 | — |
case-07 | pass→fail | 22,104 | 3,122 | -86% | 1 | 1 | 0% | 3,304 | 364 | -89% | 0 | 0 | — |
case-08 | fail→pass | 20,370 | 15,956 | -22% | 1 | 1 | 0% | 2,685 | 1,889 | -30% | 0 | 0 | — |
case-22 | pass→pass | 9,541 | 6,930 | -27% | 1 | 1 | 0% | 1,500 | 1,320 | -12% | 0 | 0 | — |
case-09 | fail→fail | 6,437 | 6,826 | +6% | 1 | 1 | 0% | 1,063 | 657 | -38% | 0 | 0 | — |
case-10 | fail→fail | 10,639 | 5,047 | -53% | 1 | 1 | 0% | 1,415 | 386 | -73% | 0 | 0 | — |
case-11 | fail→fail | 19,017 | 4,593 | -76% | 1 | 1 | 0% | 2,885 | 814 | -72% | 0 | 0 | — |
case-12 | fail→pass | 10,614 | 3,766 | -65% | 1 | 1 | 0% | 1,583 | 749 | -53% | 0 | 0 | — |
case-13 | fail→pass | 15,074 | 12,287 | -18% | 1 | 1 | 0% | 2,602 | 2,244 | -14% | 0 | 0 | — |
case-14 | fail→pass | 16,874 | 14,148 | -16% | 1 | 1 | 0% | 2,424 | 2,208 | -9% | 0 | 0 | — |
case-15 | fail→pass | 12,988 | 11,598 | -11% | 1 | 1 | 0% | 1,959 | 2,170 | +11% | 0 | 0 | — |
case-16 | fail→pass | 13,619 | 4,875 | -64% | 1 | 1 | 0% | 2,000 | 876 | -56% | 0 | 0 | — |
case-17 | fail→pass | 13,998 | 5,577 | -60% | 1 | 1 | 0% | 2,154 | 1,017 | -53% | 0 | 0 | — |
case-18 | fail→pass | 11,387 | 3,218 | -72% | 1 | 1 | 0% | 1,785 | 671 | -62% | 0 | 0 | — |
case-19 | fail→pass | 15,126 | 3,508 | -77% | 1 | 1 | 0% | 2,208 | 695 | -69% | 0 | 0 | — |
case-20 | pass→fail | 9,593 | 8,416 | -12% | 1 | 1 | 0% | 1,559 | 1,528 | -2% | 0 | 0 | — |
case-21 | fail→fail | 4,001 | 6,389 | +60% | 1 | 1 | 0% | 640 | 1,140 | +78% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +41 percentage points is the difference between those two pass rates over the 17 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.