Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Marks project context as unconfigured until project-inject materializes domain constraints, phase, conventions, and external-resource access in this file. Use when checking project context before the initial project-inject run.
.claude/skills/shinpr-project-context/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -2% | 0% |
ステータス: 未設定。
/project-injectを一度実行し、プロジェクト固有のコンテキストを収集して、この未設定状態の本文を置き換える。設定済みの本文には、少なくとも## プロジェクト概要セクションが含まれる。コマンドを利用できない、またはヒアリングを完了できない場合は、プロジェクトコンテキストが不明であることと、不足しているプロジェクト判断を具体的に報告し、推測で補わない。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | fail→pass | 12,873 | 12,195 | -5% | 1 | 1 | 0% | 2,164 | 1,575 | -27% | 0 | 0 | — |
case-01 | pass→pass | 9,216 | 8,169 | -11% | 1 | 1 | 0% | 1,571 | 1,091 | -31% | 0 | 0 | — |
case-02 | fail→pass | 21,179 | 7,963 | -62% | 1 | 1 | 0% | 3,071 | 1,494 | -51% | 0 | 0 | — |
case-03 | fail→pass | 12,549 | 6,981 | -44% | 1 | 1 | 0% | 2,070 | 1,250 | -40% | 0 | 0 | — |
case-04 | fail→fail | 11,467 | 11,912 | +4% | 1 | 1 | 0% | 1,861 | 1,964 | +6% | 0 | 0 | — |
case-11 | fail→pass | 18,947 | 10,335 | -45% | 1 | 1 | 0% | 2,885 | 1,686 | -42% | 0 | 0 | — |
case-05 | fail→pass | 7,392 | 6,657 | -10% | 1 | 1 | 0% | 1,175 | 1,146 | -2% | 0 | 0 | — |
case-06 | fail→fail | 15,675 | 12,422 | -21% | 1 | 1 | 0% | 2,530 | 2,149 | -15% | 0 | 0 | — |
case-07 | fail→pass | 10,315 | 9,990 | -3% | 1 | 1 | 0% | 1,817 | 1,651 | -9% | 0 | 0 | — |
case-08 | fail→fail | 20,711 | 20,155 | -3% | 1 | 1 | 0% | 3,562 | 3,019 | -15% | 0 | 0 | — |
case-09 | fail→fail | 19,837 | 24,094 | +21% | 1 | 1 | 0% | 3,755 | 3,281 | -13% | 0 | 0 | — |
case-12 | fail→fail | 17,186 | 16,425 | -4% | 1 | 1 | 0% | 3,042 | 2,595 | -15% | 0 | 0 | — |
case-13 | fail→pass | 9,856 | 7,372 | -25% | 1 | 1 | 0% | 1,377 | 1,242 | -10% | 0 | 0 | — |
case-14 | fail→pass | 6,977 | 3,558 | -49% | 1 | 1 | 0% | 960 | 500 | -48% | 0 | 0 | — |
case-15 | fail→pass | 14,701 | 8,311 | -43% | 1 | 1 | 0% | 2,115 | 1,285 | -39% | 0 | 0 | — |
case-16 | fail→pass | 18,474 | 31,740 | +72% | 1 | 1 | 0% | 3,122 | 2,559 | -18% | 0 | 0 | — |
case-17 | fail→pass | 16,690 | 8,196 | -51% | 1 | 1 | 0% | 1,951 | 1,221 | -37% | 0 | 0 | — |
case-18 | fail→fail | 7,400 | 4,553 | -38% | 1 | 1 | 0% | 982 | 616 | -37% | 0 | 0 | — |
case-19 | fail→pass | 39,257 | 4,560 | -88% | 1 | 1 | 0% | 1,189 | 988 | -17% | 0 | 0 | — |
case-20 | pass→pass | 4,269 | 4,074 | -5% | 1 | 1 | 0% | 572 | 717 | +25% | 0 | 0 | — |
case-21 | pass→pass | 4,193 | 5,227 | +25% | 1 | 1 | 0% | 568 | 877 | +54% | 0 | 0 | — |
case-22 | pass→pass | 18,013 | 17,982 | -0% | 1 | 1 | 0% | 2,686 | 2,640 | -2% | 0 | 0 | — |
case-23 | fail→fail | 16,993 | 19,031 | +12% | 1 | 1 | 0% | 2,426 | 2,955 | +22% | 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. 23 cases were attempted. The headline lift of +52 percentage points is the difference between those two pass rates over the 23 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.