Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →ai-agent-campのローカル環境セットアップを確認するスキル。 「セットアップ確認」「環境チェック」「初期設定できてる？」「インストール確認」「依存関係チェック」等のリクエストで発動。
.claude/skills/minicoohei-aiagent-check-setup/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -65% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -52% | 0% |
「セットアップ確認」「環境チェック」「初期設定」「インストール確認」
Use this skill to confirm the learner can safely start the course with Codex.
git --versionnode --versionnpm --versionpython3 --versionclaude --version and cursor --version only if the user wants cross-tool parity.env or credential-store setup without printing secrets.git/hooks/pre-commit after bash scripts/install_hooks.shdocs/codex-safety.md.courses/aiagent/lesson02-setup/ch01-environment/practice/checklist.md（このパスが存在しない場合はスキップ）.aiagent-lesson-runner with the next lesson id.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→fail | 8,068 | 5,953 | -26% | 1 | 1 | 0% | 1,149 | 869 | -24% | 0 | 0 | — |
case-02 | fail→fail | 11,608 | 2,496 | -78% | 1 | 1 | 0% | 1,919 | 629 | -67% | 0 | 0 | — |
case-01 | fail→fail | 12,308 | 3,656 | -70% | 1 | 1 | 0% | 2,046 | 858 | -58% | 0 | 0 | — |
case-04 | pass→fail | 10,964 | 3,285 | -70% | 1 | 1 | 0% | 1,818 | 780 | -57% | 0 | 0 | — |
case-05 | pass→pass | 13,818 | 8,180 | -41% | 1 | 1 | 0% | 2,289 | 1,077 | -53% | 0 | 0 | — |
case-06 | fail→pass | 10,721 | 6,034 | -44% | 1 | 1 | 0% | 1,665 | 1,063 | -36% | 0 | 0 | — |
case-07 | pass→fail | 11,119 | 6,238 | -44% | 1 | 1 | 0% | 1,871 | 1,054 | -44% | 0 | 0 | — |
case-08 | fail→pass | 8,633 | 3,402 | -61% | 1 | 1 | 0% | 1,290 | 719 | -44% | 0 | 0 | — |
case-09 | fail→pass | 13,464 | 6,190 | -54% | 1 | 1 | 0% | 2,073 | 1,143 | -45% | 0 | 0 | — |
case-10 | fail→pass | 14,932 | 4,131 | -72% | 1 | 1 | 0% | 2,383 | 838 | -65% | 0 | 0 | — |
case-11 | fail→pass | 15,670 | 5,326 | -66% | 1 | 1 | 0% | 2,225 | 1,061 | -52% | 0 | 0 | — |
case-12 | pass→pass | 6,193 | 5,187 | -16% | 1 | 1 | 0% | 1,010 | 772 | -24% | 0 | 0 | — |
case-13 | pass→pass | 12,208 | 3,951 | -68% | 1 | 1 | 0% | 2,202 | 841 | -62% | 0 | 0 | — |
case-14 | pass→pass | 9,255 | 2,291 | -75% | 1 | 1 | 0% | 1,553 | 540 | -65% | 0 | 0 | — |
case-15 | fail→pass | 10,667 | 1,671 | -84% | 1 | 1 | 0% | 1,607 | 457 | -72% | 0 | 0 | — |
case-16 | pass→pass | 9,062 | 3,915 | -57% | 1 | 1 | 0% | 1,382 | 850 | -38% | 0 | 0 | — |
case-17 | fail→pass | 6,879 | 1,801 | -74% | 1 | 1 | 0% | 965 | 469 | -51% | 0 | 0 | — |
case-18 | fail→pass | 12,419 | 4,343 | -65% | 1 | 1 | 0% | 1,805 | 895 | -50% | 0 | 0 | — |
case-19 | fail→pass | 14,491 | 6,261 | -57% | 1 | 1 | 0% | 2,045 | 1,239 | -39% | 0 | 0 | — |
case-20 | pass→pass | 7,548 | 5,706 | -24% | 1 | 1 | 0% | 1,236 | 1,038 | -16% | 0 | 0 | — |
case-21 | pass→pass | 20,187 | 11,561 | -43% | 1 | 1 | 0% | 3,032 | 2,272 | -25% | 0 | 0 | — |
case-22 | fail→pass | 12,870 | 6,875 | -47% | 1 | 1 | 0% | 2,372 | 1,400 | -41% | 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 +36 percentage points is the difference between those two pass rates over the 22 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.