Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Gemini Image Generation API でテキストから図解・インフォグラフィック・ダイアグラムを生成するスキル。 「図を作って」「インフォグラフィック生成」「プロセスを図解して」等のリクエストで発動。
.claude/skills/minicoohei-diagram-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -82% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -58% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -24% | 0% |
Generate visual diagrams and infographics from text descriptions.
bashpython scripts/generate_diagram.py "{topic}" --style "{style}" --aspect_ratio "{ratio}"
| Parameter | Required | Default | Description | |-----------|----------|---------|-------------| | topic | Yes | - | Topic or text to visualize | | --style | No | colorful_infographic | Visual style | | --aspect_ratio | No | 16:9 | Output ratio | | --output | No | auto | Output path |
| Style | Description | |-------|-------------| | colorful_infographic | Vibrant icons, organized layout (default) | | sketch | Hand-drawn, pencil texture | | minimalist | Clean, simple geometric shapes | | photorealistic | High-quality photo look | | claymation | 3D clay texture, playful | | pixel_art | Retro game style |
16:9, 1:1, 4:3, 3:4, 9:16, 21:9
bash# Basic diagram python scripts/generate_diagram.py "How photosynthesis works" # Minimalist style python scripts/generate_diagram.py "Machine learning pipeline" --style minimalist # Portrait format python scripts/generate_diagram.py "Company org chart" --aspect_ratio 9:16 # From long text python scripts/generate_diagram.py "Gemini is a multimodal AI model that can understand text, images, and code..." --style colorful_infographic
テキストのトピックや説明文から、Gemini Image Generation APIを使ってインフォグラフィック・図解・ビジュアルを自動生成するスキルです。プレゼン資料やドキュメントの視覚的補助に最適。
| エラー | 解決方法 | |--------|---------| | API key not found | GEMINI_API_KEY または GOOGLE_API_KEY を環境変数に設定 | | Image generation returned empty | トピックが短すぎる可能性。より詳しい説明を指定 |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 24,195 | 3,049 | -87% | 1 | 1 | 0% | 5,277 | 949 | -82% | 0 | 0 | — |
case-02 | fail→fail | 11,313 | 7,086 | -37% | 1 | 1 | 0% | 1,656 | 913 | -45% | 0 | 0 | — |
case-03 | fail→fail | 17,246 | 7,194 | -58% | 1 | 1 | 0% | 3,279 | 857 | -74% | 0 | 0 | — |
case-04 | fail→pass | 14,869 | 3,329 | -78% | 1 | 1 | 0% | 2,372 | 993 | -58% | 0 | 0 | — |
case-05 | fail→pass | 8,704 | 2,312 | -73% | 1 | 1 | 0% | 1,383 | 989 | -28% | 0 | 0 | — |
case-19 | fail→pass | 8,760 | 1,528 | -83% | 1 | 1 | 0% | 1,356 | 742 | -45% | 0 | 0 | — |
case-06 | fail→pass | 7,608 | 1,886 | -75% | 1 | 1 | 0% | 1,088 | 832 | -24% | 0 | 0 | — |
case-07 | pass→pass | 6,147 | 3,140 | -49% | 1 | 1 | 0% | 941 | 939 | -0% | 0 | 0 | — |
case-08 | fail→pass | 3,746 | 2,745 | -27% | 1 | 1 | 0% | 541 | 894 | +65% | 0 | 0 | — |
case-09 | fail→pass | 10,108 | 2,120 | -79% | 1 | 1 | 0% | 1,833 | 875 | -52% | 0 | 0 | — |
case-10 | fail→pass | 3,829 | 2,113 | -45% | 1 | 1 | 0% | 539 | 908 | +68% | 0 | 0 | — |
case-11 | pass→pass | 5,754 | 2,250 | -61% | 1 | 1 | 0% | 842 | 844 | +0% | 0 | 0 | — |
case-12 | fail→pass | 9,547 | 1,611 | -83% | 1 | 1 | 0% | 1,588 | 810 | -49% | 0 | 0 | — |
case-13 | fail→pass | 3,758 | 1,615 | -57% | 1 | 1 | 0% | 537 | 805 | +50% | 0 | 0 | — |
case-14 | fail→pass | 13,759 | 8,392 | -39% | 1 | 1 | 0% | 2,183 | 1,933 | -11% | 0 | 0 | — |
case-15 | pass→pass | 8,034 | 3,554 | -56% | 1 | 1 | 0% | 1,431 | 1,139 | -20% | 0 | 0 | — |
case-16 | fail→pass | 9,382 | 3,443 | -63% | 1 | 1 | 0% | 1,224 | 945 | -23% | 0 | 0 | — |
case-17 | fail→pass | 11,126 | 4,607 | -59% | 1 | 1 | 0% | 1,825 | 1,291 | -29% | 0 | 0 | — |
case-18 | fail→pass | 8,695 | 2,043 | -77% | 1 | 1 | 0% | 1,363 | 833 | -39% | 0 | 0 | — |
case-20 | fail→pass | 4,915 | 1,785 | -64% | 1 | 1 | 0% | 620 | 774 | +25% | 0 | 0 | — |
case-21 | pass→pass | 5,403 | 6,315 | +17% | 1 | 1 | 0% | 1,036 | 1,717 | +66% | 0 | 0 | — |
case-22 | pass→pass | 4,318 | 4,438 | +3% | 1 | 1 | 0% | 791 | 1,441 | +82% | 0 | 0 | — |
case-23 | pass→pass | 10,639 | 8,828 | -17% | 1 | 1 | 0% | 1,812 | 2,117 | +17% | 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, and 21 counted toward the lift figure. The other 2 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 +65 percentage points is the difference between those two pass rates over the 21 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.