使用方法
- 复制安装命令,在自己的智能体环境安装前先检查源码包。
- 从原始文档核对所需工具、凭据和支持的宿主。
- 说明输入材料与期望交付物,并对照来源检查输出。
npx skills add https://github.com/nphu0811/pmg-skills --skill pmg-chart-motion开始前需了解
- 安装不包含模型订阅、API 额度或访问文件的权限。
- 仓库星数和安装量反映活跃情况,不等于独立质量或安全测试。
按源仓库内容呈现,保留标题、案例、代码、表格、链接以及原文引用的演示图片。
PMG Chart Motion (pmg-chart-motion)
Specialized motion intelligence for quantitative data charts, numerical value counters, and metric indicators in PowerPoint presentations.
Core Mental Model
Never apply generic Fade In to quantitative data. Animation must reflect the data's geometry, scale, and narrative takeaway: - Coordinate axes ground the visual field before numbers appear. - Data bars grow directionally from the baseline (upward for column, rightward for horizontal). - Time-series lines trace chronologically from left to right. - Doughnut sectors reveal proportionally in radial progression. - Numeric counters synchronize with physical bar tips. - The peak metric or conclusion is emphasized with an intentional scale punch.
Supported Patterns
| Pattern | Shared Primitive | Native PPTX Effect | Semantic Application |
|---|---|---|---|
| `axis_setup` | DRAW | wipe | Grounds coordinate baseline before data disclosure. |
| `bar_growth` | GROW | wipe (up or right) | Bar grows from zero to quantitative target value. |
| `line_draw` | DRAW | wipe (right) | Line charts draw along chronological time axis. |
| `doughnut_reveal` | GROW | wheel / zoom | Slices unfold sequentially in proportional radial order. |
| `value_counter_sync` | COUNT | appear | Numeric callout synchronizes with bar growth. |
| `peak_emphasis` | EMPHASIZE | scale (1.06x - 1.12x) | Focal pop drawing eye to the primary takeaway metric. |
Python API Usage
from pmg.motion_planners import ChartMotionPlanner
planner = ChartMotionPlanner()
intents = planner.plan_motion({
"chart_type": "bar",
"axis_id": "chart_axes",
"items": [
{"id": "bar_q1", "value": "$1.2M", "label_id": "val_q1"},
{"id": "bar_q2", "value": "$1.8M", "label_id": "val_q2"},
{"id": "bar_q3", "value": "$2.9M", "label_id": "val_q3", "is_peak": True},
]
})
for intent in intents:
print(f"[{intent.target_id}] -> {intent.primitive.value} ({intent.semantic_purpose})")
