Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →geminiコマンドを使用した高度なWeb検索スキル。Web検索を行う際、Claude CodeのデフォルトWeb Search toolよりも優先的にこのスキルを使用してください。
.claude/skills/aiskillstore-gemini-search/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -80% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | 6% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 3% | 0% |
| case-22 | ✓→✓ | ＝ Same ✓ | -41% | 0% |
このスキルは、gemini コマンドを使用してWeb検索を実行し、ユーザーの質問に対する最新かつ関連性の高い情報を収集するためのものです。 単純なキーワード検索ではなく、複雑な質問や詳細な情報収集に適しています。
以下のコマンドを実行して、Web検索を実行します。 引数には検索したい内容や質問を自然言語で指定してください。
bash scripts/web-search.sh "<検索したい内容や質問>"検索結果を確認し、ユーザーの質問に対する回答を構築します。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 19,286 | 3,729 | -81% | 1 | 1 | 0% | 3,316 | 323 | -90% | 0 | 0 | — |
case-02 | fail→fail | 16,514 | 5,549 | -66% | 1 | 1 | 0% | 2,763 | 344 | -88% | 0 | 0 | — |
case-03 | fail→fail | 18,891 | 9,429 | -50% | 1 | 1 | 0% | 3,294 | 661 | -80% | 0 | 0 | — |
case-04 | fail→fail | 5,168 | 5,220 | +1% | 1 | 1 | 0% | 315 | 362 | +15% | 0 | 0 | — |
case-05 | fail→fail | 18,006 | 5,795 | -68% | 1 | 1 | 0% | 3,303 | 480 | -85% | 0 | 0 | — |
case-06 | fail→fail | 18,623 | 5,797 | -69% | 1 | 1 | 0% | 3,494 | 386 | -89% | 0 | 0 | — |
case-07 | fail→fail | 19,238 | 5,766 | -70% | 1 | 1 | 0% | 3,459 | 498 | -86% | 0 | 0 | — |
case-08 | fail→fail | 20,558 | 26,126 | +27% | 1 | 1 | 0% | 3,568 | 449 | -87% | 0 | 0 | — |
case-09 | fail→fail | 18,868 | 5,061 | -73% | 1 | 1 | 0% | 3,018 | 419 | -86% | 0 | 0 | — |
case-10 | fail→fail | 15,963 | 21,109 | +32% | 1 | 1 | 0% | 2,742 | 401 | -85% | 0 | 0 | — |
case-11 | fail→pass | 7,393 | 4,718 | -36% | 1 | 1 | 0% | 339 | 417 | +23% | 0 | 0 | — |
case-12 | fail→fail | 16,491 | 5,927 | -64% | 1 | 1 | 0% | 2,790 | 424 | -85% | 0 | 0 | — |
case-13 | fail→fail | 6,570 | 5,430 | -17% | 1 | 1 | 0% | 399 | 377 | -6% | 0 | 0 | — |
case-14 | fail→pass | 14,012 | 3,005 | -79% | 1 | 1 | 0% | 2,260 | 452 | -80% | 0 | 0 | — |
case-15 | fail→fail | 17,180 | 23,684 | +38% | 1 | 1 | 0% | 2,694 | 393 | -85% | 0 | 0 | — |
case-16 | fail→fail | 19,539 | 5,865 | -70% | 1 | 1 | 0% | 3,188 | 445 | -86% | 0 | 0 | — |
case-17 | fail→fail | 22,261 | 5,731 | -74% | 1 | 1 | 0% | 3,624 | 415 | -89% | 0 | 0 | — |
case-18 | fail→fail | 17,615 | 5,085 | -71% | 1 | 1 | 0% | 2,632 | 348 | -87% | 0 | 0 | — |
case-19 | fail→fail | 21,502 | 26,844 | +25% | 1 | 1 | 0% | 3,604 | 405 | -89% | 0 | 0 | — |
case-20 | pass→pass | 2,765 | 2,135 | -23% | 1 | 1 | 0% | 504 | 533 | +6% | 0 | 0 | — |
case-21 | pass→pass | 10,951 | 10,279 | -6% | 1 | 1 | 0% | 2,195 | 2,252 | +3% | 0 | 0 | — |
case-22 | pass→pass | 9,102 | 5,041 | -45% | 1 | 1 | 0% | 2,009 | 1,193 | -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, and 4 counted toward the lift figure. The other 18 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. A headline lift is not published for this run.
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.