Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

沁言 Nature 科研绘图 · leonchaox | DecimalAI

scienceby leonchaox

沁言 Nature 科研绘图

leonchaox/qinyan-nature-figures

Source leonchaox/qinyan-academic-skills ↗License MIT

面向 Nature Portfolio 与高影响力期刊的证据驱动科研绘图技能。用于从原始或汇总数据设计单图与多面板 figure、选择合适图形语法、编写 Python/R 绘图代码、重绘现有图件、生成机制示意图草案、撰写图注并导出可编辑 SVG/PDF 与高分辨率 TIFF/PNG；同时检查数据完整性、颜色可访问性、统计标注和最终尺寸可读性。触发场景包括 Nature 绘图、科研作图、论文配图、scientific figure、publication plot、multi-panel figure、graphical abstract、机制图、图形摘要和 figure audit。

SkillScore82Maintained

Automatic

Research pack

Last verified 8/15/2026

What it changes

Passed 18 of 22 test cases with the skill, and 13 of 22 without it.

Without skill59%

With skill82%

Checks it passed that the model alone missed

Includes input rows in the integrity log.
Includes Panel map in the contract items.
Includes output rows in the integrity log.

and 7 more in the full report

6 fixed12 passed either way3 still failing1 broke

Pass-rate lift: +23 pp
Tokens: +38%
Agent turns: 0%
Response time: -9%

Model gemini-3.6-flashMeasured 8/15/2026v118 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/leonchaox-qinyan-nature-figures/SKILL.md https://app.decimal.ai/s/leonchaox-qinyan-nature-figures/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add leonchaox-qinyan-nature-figures, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/leonchaox-qinyan-nature-figures/ and pins it in .decimal/skills.lock

decimalai skills add leonchaox-qinyan-nature-figures

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/leonchaox-qinyan-nature-figures/SKILL.md https://app.decimal.ai/s/leonchaox-qinyan-nature-figures/SKILL.md

Raw file: /s/leonchaox-qinyan-nature-figures/SKILL.md ↗ · machine-readable summary at /s/leonchaox-qinyan-nature-figures.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 6

Prompt

We finished writing our matplotlib script `plot_survival.py` and exported `figure1.svg`. What command line tool and script should be executed to perform the automated preflight check?

What was checked, without and with the skill
Checked	Without skill	With skill
Specifies running python scripts/figure_preflight.py with source and artifact flags.	Missed	Passed

case-12 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md84 lines5 filesincludes scripts

沁言 Nature 科研绘图

先定义图要证明什么，再决定画什么。期刊级图件是证据结构、视觉层级、数据诚信和可复现导出的共同产物。

路由

定量图件：使用 Python（matplotlib/seaborn）或 R（ggplot2/patchwork/ComplexHeatmap）。
机制图或图形摘要：先建立概念与关系清单，再使用矢量工具或可用的图像生成能力制作草案；不得用 AI 图替代定量证据。
已有图件审查：同时检查图源代码、最终导出和最终版面尺寸，不能只看屏幕截图。

优先服从用户现有语言与项目栈。用户未指定且不存在项目约束时，默认使用 Python，并在交付中说明；只有选择会显著影响复现或协作时才询问。

图件契约

绘图前写出：

Conclusion：读者看完图后应能复述的一句话。
Evidence hierarchy：主证据、支持证据、对照与边界。
Panel map：每个面板的任务、数据和与其他面板的关系。
Data contract：变量、单位、独立样本、缺失、排除和变换规则。
Statistics contract：估计量、误差、检验、校正、n 与配对/重复结构。
Export contract：栏宽、目标尺寸、字体、矢量/栅格、分辨率和 source data。

详细模板读取 references/figure-contract.md。

执行流程

审计数据。 保留输入行数、排除规则、变换和聚合前后计数；不得静默删除异常或缺失值。
选择图形。 根据科学问题、变量类型和实验层级选图，不按“看起来像顶刊”选图。读取 references/chart-selection.md。
规划版面。 先安排主面板与阅读顺序，再写绘图代码；面板数量服务论证，不追求填满页面。
编码与导出。 固定随机种子、字体、尺寸、颜色、排序和导出参数；保留可运行源码。
写图注。 说明样本、n、中心量、误差、检验、校正、符号和缩写，使图注可独立理解。
自动预检。 执行 python scripts/figure_preflight.py <source.py|source.R> --artifact <figure.svg> ...。
视觉核验。 在最终印刷尺寸检查标签、图例、线宽、遮挡、色盲可辨识、面板一致性和缩放后的栅格清晰度。
交付溯源包。 提供源码、导出文件、source data、参数说明、排除记录和剩余风险。

视觉与导出标准读取 references/visual-standards.md。

默认交付

Track record

Benchmark

18 / 22

test cases passed, vs 13 / 22 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

leonchaox / qinyan-academic-skills

skills/%E6%B2%81%E8%A8%80%E5%AD%A6%E6%9C%AFskills/qinyan-nature-figures/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1Bundled 4 files · scripts

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

沁言 Nature 科研绘图

路由

图件契约

执行流程

默认交付

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills