Install any skill in seconds. Free to start, no credit card required.
Get Started Free →将 JSON 格式的 PPT 内容转换为标准的 .pptx 文件。使用 python-pptx 库,支持多种布局、图表、表格和样式。与 ppt-generator Skill 完全协同,可作为独立使用或与其他 PPT Skill 配合。
.claude/skills/anbeime-pptx-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -56% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 19% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 114% | 0% |
python-pptx>=1.0.2 pillow>=9.0.0 openpyxl>=3.1.0
从以下来源接收 JSON 数据:
调用 scripts/json_validator.py 验证 JSON 格式:
bashpython scripts/json_validator.py --input ./ppt_data.json
验证内容包括:
调用 scripts/pptx_builder.py 生成 .pptx 文件:
bashpython scripts/pptx_builder.py \ --input ./ppt_data.json \ --style assets/styles/modern.json \ --output ./presentation.pptx
核心功能:
调用 scripts/pptx_validator.py 验证生成的 .pptx 文件:
bashpython scripts/pptx_validator.py --input ./presentation.pptx
验证内容包括:
使用自定义风格配置:
bashpython scripts/pptx_builder.py \ --input ./ppt_data.json \ --style ./custom_style.json \ --output ./presentation.pptx
批量处理多个 JSON 文件:
bashpython scripts/pptx_builder.py \ --input-dir ./json_files/ \ --style assets/styles/modern.json \ --output-dir ./pptx_files/
基于模板生成 PPTX:
bashpython scripts/pptx_builder.py \ --input ./ppt_data.json \ --template assets/templates/business_template.pptx \ --output ./presentation.pptx
用户请求:"生成一个 PPT 文件"
┌─────────────────────────────────────────────────────────┐
│ ppt-generator Skill │
├─────────────────────────────────────────────────────────┤
│ 1. 主题分析师:分析主题 │
│ 2. 模板设计师:推荐布局 │
│ 3. 内容策划师:规划内容结构 │
│ 4. 文本创作者:撰写内容 │
│ 5. 视觉设计师:提供配图建议 │
│ 6. 优化编辑师:优化文本 │
│ 7. PPT 构建师:生成 JSON 数据 │
└─────────────────────────────────────────────────────────┘
│
▼ ppt_data.json
┌─────────────────────────────────────────────────────────┐
│ pptx-generator Skill │
├─────────────────────────────────────────────────────────┤
│ 1. JSON 验证(json_validator.py) │
│ 2. PPTX 构建(pptx_builder.py) │
│ - 创建演示文稿 │
│ - 添加幻灯片 │
│ - 应用布局和样式 │
│ - 添加内容(文本、图片、图表、表格) │
│ 3. PPTX 验证(pptx_validator.py) │
│ 4. 输出 .pptx 文件 │
└─────────────────────────────────────────────────────────┘
│
▼ presentation.pptx用户请求:"生成 PPT 文件和图片"
ppt-generator → pptx-generator → nanobanana-ppt-visualizer
↓ ↓ ↓
JSON 数据 .pptx 文件 图片 + HTML 播放器1. ppt-generator → JSON 数据
2. pptx-generator → .pptx 文件
3. nanobanana-ppt-visualizer → 图片 + HTML 播放器
4. ppt-roadshow-generator → 视频
5. remotion-video-enhancer → 增强动画python scripts/pptx_builder.py --input ./ppt_data.json --output ./presentation.pptxpython scripts/pptx_builder.py --input ./ppt_data.json --style ./my_style.json --output ./presentation.pptxpython scripts/pptx_builder.py --input-dir ./json_files/ --output-dir ./pptx_files/| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 6,949 | 3,137 | -55% | 1 | 1 | 0% | 1,361 | 2,526 | +86% | 0 | 0 | — |
case-01 | fail→pass | 26,219 | 4,811 | -82% | 1 | 1 | 0% | 6,194 | 2,909 | -53% | 0 | 0 | — |
case-02 | fail→pass | 30,071 | 7,623 | -75% | 1 | 1 | 0% | 6,192 | 3,174 | -49% | 0 | 0 | — |
case-03 | fail→pass | 25,173 | 4,168 | -83% | 1 | 1 | 0% | 6,200 | 2,752 | -56% | 0 | 0 | — |
case-04 | fail→fail | 15,977 | 14,666 | -8% | 1 | 1 | 0% | 3,618 | 4,904 | +36% | 0 | 0 | — |
case-05 | fail→pass | 16,118 | 8,534 | -47% | 1 | 1 | 0% | 3,002 | 3,562 | +19% | 0 | 0 | — |
case-06 | fail→fail | 16,488 | 21,576 | +31% | 1 | 1 | 0% | 3,490 | 5,739 | +64% | 0 | 0 | — |
case-07 | fail→pass | 5,994 | 3,693 | -38% | 1 | 1 | 0% | 1,116 | 2,393 | +114% | 0 | 0 | — |
case-08 | fail→pass | 7,430 | 3,316 | -55% | 1 | 1 | 0% | 1,502 | 2,583 | +72% | 0 | 0 | — |
case-09 | fail→pass | 11,966 | 2,825 | -76% | 1 | 1 | 0% | 1,738 | 2,445 | +41% | 0 | 0 | — |
case-10 | fail→pass | 8,785 | 2,598 | -70% | 1 | 1 | 0% | 1,495 | 2,427 | +62% | 0 | 0 | — |
case-11 | fail→pass | 8,725 | 3,763 | -57% | 1 | 1 | 0% | 1,599 | 2,481 | +55% | 0 | 0 | — |
case-12 | pass→pass | 9,721 | 2,690 | -72% | 1 | 1 | 0% | 1,843 | 2,391 | +30% | 0 | 0 | — |
case-13 | pass→pass | 9,522 | 8,689 | -9% | 1 | 1 | 0% | 1,798 | 3,564 | +98% | 0 | 0 | — |
case-14 | pass→pass | 13,221 | 9,286 | -30% | 1 | 1 | 0% | 2,081 | 3,464 | +66% | 0 | 0 | — |
case-15 | fail→pass | 5,818 | 1,800 | -69% | 1 | 1 | 0% | 1,034 | 2,232 | +116% | 0 | 0 | — |
case-16 | fail→pass | 8,488 | 2,695 | -68% | 1 | 1 | 0% | 1,682 | 2,383 | +42% | 0 | 0 | — |
case-17 | fail→pass | 8,946 | 3,580 | -60% | 1 | 1 | 0% | 1,471 | 2,624 | +78% | 0 | 0 | — |
case-18 | fail→pass | 8,889 | 2,724 | -69% | 1 | 1 | 0% | 1,531 | 2,447 | +60% | 0 | 0 | — |
case-19 | fail→pass | 14,416 | 8,242 | -43% | 1 | 1 | 0% | 2,255 | 3,373 | +50% | 0 | 0 | — |
case-21 | fail→pass | 13,548 | 6,846 | -49% | 1 | 1 | 0% | 2,078 | 2,475 | +19% | 0 | 0 | — |
case-22 | fail→pass | 14,914 | 2,456 | -84% | 1 | 1 | 0% | 2,499 | 2,432 | -3% | 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.