Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →PPT 视觉增强工具，支持多种风格渲染、交互式播放器生成和视频合成。可与 ppt-generator Skill 协同工作，实现从内容规划到视觉呈现的完整流程。
.claude/skills/anbeime-nanobanana-ppt-visualizer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -5% | 0% |
pillow>=9.0.0 python-dotenv>=0.19.0
bash # Ubuntu/Debian sudo apt-get install ffmpeg
# macOS brew install ffmpeg
assets/styles/ 目录，列出可用风格：gradient-glass.md：渐变毛玻璃风格（科技感、商务）vector-illustration.md：矢量插画风格（温暖、教育）scripts/video_materials.py 管理视频素材scripts/generate_viewer.py 生成 HTML 播放器：scripts/video_composer.py：当用户直接提供 PPT 内容（JSON 格式）时：
用户请求："生成一个关于 AI 产品的 PPT"
┌─────────────────────────────────────────────────────────┐
│ ppt-generator Skill │
├─────────────────────────────────────────────────────────┤
│ 1. 主题分析师：分析主题，生成大纲 │
│ 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 │
└─────────────────────────────────────────────────────────┘
│
▼ 输出 JSON
┌─────────────────────────────────────────────────────────┐
│ nanobanana-ppt-visualizer Skill │
├─────────────────────────────────────────────────────────┤
│ 1. 接收 JSON 数据 │
│ 2. 选择视觉风格 │
│ 3. 生成图片（使用智能体能力） │
│ 4. 生成 HTML 播放器 │
│ 5. 可选：视频合成 │
└─────────────────────────────────────────────────────────┘
│
▼ 输出
交互式播放器 + 完整视频（可选）ppt-generator 输出的 JSON 格式与 nanobanana-ppt-visualizer 完全兼容：
json{ "metadata": { "title": "演示文稿标题", "author": "作者姓名" }, "slides": [ { "layout": "TitleSlide", "title": "封面标题", "content": ["副标题"], "notes": "备注" } ] }
python scripts/generate_viewer.py --input ./ppt_data.json --style gradient-glasspython scripts/video_composer.py --output ./full_ppt_video.mp4| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | fail→pass | 16,000 | 5,284 | -67% | 1 | 1 | 0% | 3,119 | 2,731 | -12% | 0 | 0 | — |
case-01 | fail→fail | 34,766 | 7,543 | -78% | 1 | 1 | 0% | 6,195 | 2,804 | -55% | 0 | 0 | — |
case-02 | fail→fail | 12,493 | 11,029 | -12% | 1 | 1 | 0% | 1,980 | 3,342 | +69% | 0 | 0 | — |
case-03 | fail→fail | 26,301 | 7,368 | -72% | 1 | 1 | 0% | 6,188 | 2,851 | -54% | 0 | 0 | — |
case-04 | fail→fail | 30,477 | 24,798 | -19% | 1 | 1 | 0% | 5,096 | 6,570 | +29% | 0 | 0 | — |
case-05 | fail→fail | 3,879 | 3,364 | -13% | 1 | 1 | 0% | 584 | 2,305 | +295% | 0 | 0 | — |
case-06 | fail→fail | 22,254 | 19,411 | -13% | 1 | 1 | 0% | 3,433 | 5,616 | +64% | 0 | 0 | — |
case-07 | fail→pass | 39,441 | 3,397 | -91% | 1 | 1 | 0% | 3,616 | 2,296 | -37% | 0 | 0 | — |
case-08 | fail→pass | 15,829 | 3,992 | -75% | 1 | 1 | 0% | 2,500 | 2,574 | +3% | 0 | 0 | — |
case-09 | fail→pass | 18,686 | 4,569 | -76% | 1 | 1 | 0% | 2,186 | 2,576 | +18% | 0 | 0 | — |
case-10 | fail→pass | 16,470 | 8,683 | -47% | 1 | 1 | 0% | 2,898 | 2,763 | -5% | 0 | 0 | — |
case-11 | pass→pass | 13,857 | 3,058 | -78% | 1 | 1 | 0% | 2,297 | 2,330 | +1% | 0 | 0 | — |
case-12 | fail→pass | 12,970 | 6,394 | -51% | 1 | 1 | 0% | 2,130 | 2,625 | +23% | 0 | 0 | — |
case-13 | pass→pass | 17,385 | 6,373 | -63% | 1 | 1 | 0% | 2,532 | 2,734 | +8% | 0 | 0 | — |
case-14 | fail→pass | 10,817 | 3,109 | -71% | 1 | 1 | 0% | 1,841 | 2,349 | +28% | 0 | 0 | — |
case-15 | fail→pass | 12,326 | 3,486 | -72% | 1 | 1 | 0% | 2,018 | 2,248 | +11% | 0 | 0 | — |
case-16 | fail→pass | 9,181 | 2,327 | -75% | 1 | 1 | 0% | 1,509 | 2,106 | +40% | 0 | 0 | — |
case-17 | pass→pass | 14,137 | 1,872 | -87% | 1 | 1 | 0% | 1,851 | 2,076 | +12% | 0 | 0 | — |
case-18 | pass→pass | 14,543 | 2,921 | -80% | 1 | 1 | 0% | 2,394 | 2,279 | -5% | 0 | 0 | — |
case-19 | fail→fail | 11,375 | 5,857 | -49% | 1 | 1 | 0% | 2,027 | 2,676 | +32% | 0 | 0 | — |
case-21 | pass→pass | 13,379 | 4,065 | -70% | 1 | 1 | 0% | 2,248 | 2,467 | +10% | 0 | 0 | — |
case-22 | fail→pass | 8,288 | 3,385 | -59% | 1 | 1 | 0% | 1,281 | 2,394 | +87% | 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 +45 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.