AI Skill Hub 强烈推荐:SCI图表生成 是一款优质的Agent工作流。AI 综合评分 8.0 分,在同类工具中表现稳健。如果你正在寻找可靠的Agent工作流解决方案,这是一个值得深入了解的选择。
AI生成SCI和Elsevier风格的科学图表
SCI图表生成 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
AI生成SCI和Elsevier风格的科学图表
SCI图表生成 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 克隆仓库 git clone https://github.com/xiao-yuling/sci-figure cd sci-figure # 查看安装说明 cat README.md # 按 README 完成环境依赖安装后即可使用
# 查看帮助 sci-figure --help # 基本运行 sci-figure [options] <input> # 详细使用说明请查阅文档 # https://github.com/xiao-yuling/sci-figure
# sci-figure 配置说明 # 查看配置选项 sci-figure --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export SCI_FIGURE_CONFIG="/path/to/config.yml"
English | 中文
sci-figure is a general-purpose AI skill for scientific figure generation. It provides reusable instructions, references, and examples that help AI agents create publication-ready SCI-style figures with an Elsevier-friendly visual language.
Project page: https://xiao-yuling.github.io/sci-figure/
sci-figure was inspired by the organization of the nature-figure skill, especially its idea of starting from a figure contract before styling. This project is not a copy of that skill: it uses its own platform-independent instructions, examples, and style rules. The defaults here are tuned for Elsevier-style manuscript figures: Times-style typography, full-box axes, outward ticks on the left and bottom only, compact panel titles, and restrained scientific palettes.
sci-figure is not tied to any specific AI platform. Use it as an AI skill by copying skill.md into an agent prompt, project memory, custom assistant, workflow template, or instruction bundle.
skill.md 作为可复用 skill 提供给 AI agent。示例提示词:
Use the sci-figure skill to create a Python scatter plot for a manuscript.
The plot should compare measured and predicted values, export SVG/PDF/TIFF,
and follow the default SCI style.
运行示例脚本:
python examples/python/scatter_example.py
python examples/python/line_example.py
python examples/python/histogram_example.py
python examples/python/bar_example.py
python examples/python/heatmap_example.py
python examples/python/violin_box_example.py
python examples/python/errorbar_example.py
python examples/python/multipanel_example.py
每个脚本都会把 PNG 预览图写入 examples/gallery/。
These previews use simulated data and the default style rules: Times-style typography, full-box axes, outward ticks on the left and bottom only, editable export settings, and neutral scientific palettes.
| Figure | Preview | Demonstrates |
|---|---|---|
| Scatter plot |  | Correlation, 1:1 reference, linear fit, compact statistics annotation |
| Line plot |  | Multi-series trend comparison, restrained legend, light guide grid |
| Histogram |  | Distribution summary, reference line, mean marker, compact axis labels |
| Grouped bar plot |  | Group comparison, uncertainty caps, restrained categorical color |
| Heatmap |  | Matrix patterns, diverging scale, compact colorbar |
| Violin and box plot |  | Distribution shape, median emphasis, group spread |
| Interval plot |  | Effect estimates, uncertainty intervals, null reference |
| Multi-panel figure |  | Integrated panel titles, mixed evidence, compact layout |
Gallery PNGs are lightweight previews for documentation. For manuscript use, regenerate editable SVG/PDF and high-resolution TIFF/PNG from the scripts.
skill.md to an AI agent as a reusable skill.Example prompt:
Use the sci-figure skill to create a Python scatter plot for a manuscript.
The plot should compare measured and predicted values, export SVG/PDF/TIFF,
and follow the default SCI style.
Run the included Python examples:
python examples/python/scatter_example.py
python examples/python/line_example.py
python examples/python/histogram_example.py
python examples/python/bar_example.py
python examples/python/heatmap_example.py
python examples/python/violin_box_example.py
python examples/python/errorbar_example.py
python examples/python/multipanel_example.py
Each script writes a PNG preview to examples/gallery/.
下面的图片都由模拟数据生成,展示默认样式规则:Times 风格字体、四边框、 仅左/下外向刻度、可编辑导出设置,以及中性的科研配色。
| 图类型 | 预览 | 展示内容 |
|---|---|---|
| 散点图 |  | 相关关系、1:1 参考线、线性拟合、紧凑统计标注 |
| 折线图 |  | 多序列趋势比较、克制图例、浅色辅助网格 |
| 直方图 |  | 分布概览、参考线、均值标记、紧凑坐标轴标签 |
| 分组柱状图 |  | 组间比较、误差帽、克制分类配色 |
| 热图 |  | 矩阵模式、发散色标、紧凑 colorbar |
| 小提琴 + 箱线图 |  | 分布形状、中位数强调、组内离散程度 |
| 区间图 |  | 效应估计、不确定性区间、零值参考线 |
| 多子图组合 |  | 集成式子图标题、混合证据、紧凑版式 |
图库中的 PNG 是 README 预览图。正式投稿时,应从脚本重新导出可编辑的 SVG/PDF 和高分辨率 TIFF/PNG。
高质量的AI技能,生成专业的科学图表
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
总体来看,SCI图表生成 是一款质量优秀的Agent工作流,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。
| 原始名称 | sci-figure |
| 原始描述 | 开源AI工作流:AI skill for scientific figure generation, tuned for SCI and Elsevier-style manu。⭐6 |
| Topics | ai-skillelsevierggplot2sci |
| GitHub | https://github.com/xiao-yuling/sci-figure |
| License | MIT |
收录时间:2026-05-30 · 更新时间:2026-05-30 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端