Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →現セッションのリアルトークン使用量と推定削減量を表示。Claude Code セッションログから直接読込 — AI 推定なし。 `/genshijin-stats` で起動。出力は mode-tracker フックが注入し、モデル自身は数値計算しない。
.claude/skills/interfacex-co-jp-genshijin-stats/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -84% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -69% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -72% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -61% | 0% |
このスキルは hooks/genshijin-stats.js が提供（hooks/genshijin-mode-tracker.js が /genshijin-stats 検出時に呼出）。フックが decision: "block" で整形済 stats を reason として返す → ユーザーは即座に数値を見る。モデル側で何もする必要なし。
--share — ツイート可能な1行サマリ--all — Lifetime 集計--since 7d / --since 24h — 期間指定 Lifetime 集計| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | fail→pass | 10,826 | 1,233 | -89% | 1 | 1 | 0% | 1,690 | 264 | -84% | 0 | 0 | — |
case-01 | fail→pass | 5,034 | 2,494 | -50% | 1 | 1 | 0% | 634 | 324 | -49% | 0 | 0 | — |
case-02 | fail→pass | 8,974 | 1,912 | -79% | 1 | 1 | 0% | 1,223 | 378 | -69% | 0 | 0 | — |
case-03 | fail→pass | 8,502 | 1,759 | -79% | 1 | 1 | 0% | 1,195 | 329 | -72% | 0 | 0 | — |
case-04 | fail→pass | 9,743 | 2,691 | -72% | 1 | 1 | 0% | 1,488 | 583 | -61% | 0 | 0 | — |
case-05 | fail→pass | 14,437 | 2,163 | -85% | 1 | 1 | 0% | 2,190 | 508 | -77% | 0 | 0 | — |
case-06 | fail→pass | 9,118 | 1,302 | -86% | 1 | 1 | 0% | 1,356 | 275 | -80% | 0 | 0 | — |
case-08 | fail→pass | 15,073 | 1,767 | -88% | 1 | 1 | 0% | 2,257 | 365 | -84% | 0 | 0 | — |
case-09 | pass→pass | 10,767 | 2,088 | -81% | 1 | 1 | 0% | 1,788 | 375 | -79% | 0 | 0 | — |
case-10 | fail→pass | 14,374 | 2,524 | -82% | 1 | 1 | 0% | 1,892 | 492 | -74% | 0 | 0 | — |
case-11 | fail→pass | 12,733 | 2,119 | -83% | 1 | 1 | 0% | 1,918 | 451 | -76% | 0 | 0 | — |
case-12 | fail→pass | 23,242 | 1,362 | -94% | 1 | 1 | 0% | 3,675 | 302 | -92% | 0 | 0 | — |
case-13 | fail→pass | 10,953 | 1,435 | -87% | 1 | 1 | 0% | 1,646 | 328 | -80% | 0 | 0 | — |
case-14 | fail→pass | 15,845 | 1,464 | -91% | 1 | 1 | 0% | 2,289 | 348 | -85% | 0 | 0 | — |
case-15 | fail→pass | 14,073 | 2,347 | -83% | 1 | 1 | 0% | 2,133 | 479 | -78% | 0 | 0 | — |
case-16 | pass→pass | 12,994 | 2,719 | -79% | 1 | 1 | 0% | 2,208 | 486 | -78% | 0 | 0 | — |
case-17 | fail→pass | 12,281 | 3,896 | -68% | 1 | 1 | 0% | 1,660 | 705 | -58% | 0 | 0 | — |
case-18 | pass→pass | 10,079 | 2,013 | -80% | 1 | 1 | 0% | 1,382 | 408 | -70% | 0 | 0 | — |
case-19 | fail→pass | 9,588 | 1,131 | -88% | 1 | 1 | 0% | 1,436 | 304 | -79% | 0 | 0 | — |
case-20 | pass→pass | 12,282 | 5,498 | -55% | 1 | 1 | 0% | 1,925 | 923 | -52% | 0 | 0 | — |
case-21 | pass→pass | 11,755 | 4,984 | -58% | 1 | 1 | 0% | 2,164 | 1,003 | -54% | 0 | 0 | — |
case-22 | pass→pass | 15,266 | 8,125 | -47% | 1 | 1 | 0% | 2,155 | 1,190 | -45% | 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. The headline lift of +73 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.