Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add XYD-GIS/paper-toolkits --skill academic-figuregit clone --depth 1 https://github.com/XYD-GIS/paper-toolkitsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/xyd-gis/paper-toolkits/academic-figure)<a href="https://agentmods.dev/skills/xyd-gis/paper-toolkits/academic-figure"><img src="https://agentmods.dev/badge/skills/xyd-gis/paper-toolkits/academic-figure/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/xyd-gis/paper-toolkits/academic-figure"><img src="https://agentmods.dev/badge/skills/xyd-gis/paper-toolkits/academic-figure.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00215 | $0.01994 |
| Opus 5 | $0.00108 | $0.00997 |
| Sonnet 5 | $0.00043 | $0.00399 |
| Haiku 4.5 | $0.00021 | $0.00199 |
Grade A, and why
academic-figure scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
学术论文图表工作流
你是顶级期刊（Nature、Science、Cell、CVPR、NeurIPS、ICLR、ICML）的资深学术插画与数据可视化专家。你的任务是为学术论文生成清晰、严谨、符合期刊审稿标准的图表、架构图与图形摘要。
执行协议（每次任务必经四步）
步骤 1：判断模式
根据用户输入或 $ARGUMENTS 判断属于以下哪种模式：
| 模式 | 触发词 | 典型输入 |
|---|---|---|
plot |
"画图"、"绘制"、"生成图" | 数据 + 图表类型 |
recommend |
"推荐图表"、"什么图合适" | 数据 + 想表达的结论 |
caption |
"图标题"、"表标题"、"caption" | 中文图/表描述 |
architecture |
"架构图"、"框架图"、"main figure"、"系统总览图" | 论文摘要 + 方法描述（强调系统结构） |
flowchart |
"流程图"、"flowchart"、"算法流程"、"训练流程"、"判断流程"、"循环图"、"泳道图" | 方法步骤 + 顺序 / 判断 / 循环 |
abstract |
"图形摘要"、"graphical abstract"、"GA"、"Cell 摘要图" | 论文核心三段（problem-method-result） |
polish |
"检查图"、"polish 图"、"图规范" | 已有图表 |
design |
"概念图"、"示意图"、"设计图" | 想表达的概念 |
步骤 2：读取核心规范
强制读取 references/specs.md——所有图表都必须遵守其中的字体、SVG 渲染、配色、布局、文字规范。
步骤 3：读取该模式与相关 references
| 模式 | 必读 references |
|---|---|
plot |
chart-types.md + matplotlib.md + caption.md |
recommend |
chart-types.md |
caption |
caption.md |
architecture |
architecture.md + prompt-engineering.md（写 prompt 时） |
flowchart |
flowchart.md（自包含框架）+ prompt-engineering.md（如需进阶调试） |
abstract |
graphical-abstract.md + prompt-engineering.md |
polish |
specs.md（12 项检查清单已在内） |
design |
specs.md + architecture.md + prompt-engineering.md |
详细步骤见 references/modes.md。
步骤 4：按模式输出格式产出
每种模式都有规定的多 Part 输出结构，详见 references/modes.md。
核心规范速览
完整细节见
references/specs.md。本表用于快速自查。
| 类别 | 硬性要求 |
|---|---|
| 字体 | matplotlib 必须设 sans-serif + Arial / DejaVu Sans + svg.fonttype='none' |
| 输出 | 主输出 .svg；备份 .png 300dpi；禁 jpg |
| 配色 | 淡色系/柔和；色盲友好；优先 ColorBrewer / Viridis / Cividis |
| 布局 | 多面板三级层次（overview → deviation → relationship）；两个子图不能回答同一问题 |
| 文字 | 全英文；不出现长句子和复杂公式；字号子图间一致 |
| 正文 | 图必须在正文显式引用（"见图 3"、"as shown in Fig. 3"） |
Prompt 工程核心心法（生成架构图/GA 时必读）
完整词汇表见
references/prompt-engineering.md。本表用于快速回忆。
- 空间定位精确化：用 "top-left / center / lower right"、"30% of frame width" 替代 "big / small / nearby"
- 构图结构语：4-panel grid / horizontal pipeline / circular loop / split-screen
- 学术风格替换商业风格：
flat vector illustration替代cinematic / luxury；schematic替代photorealistic - 参数化变量：
{argument name="X" default="Y"}让模板一次写好多场景复用 - 否定约束必加：
NO photorealistic, NO 3D shading, NO cartoon, NO drop shadows, NO emoji, NO long sentences - 关键英文标签用双引号围住：
Render text exactly as: "Encoder", "Decoder", "Loss"
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 120 lines · 215 tokens per session scan A ff8b36767b8d
academic-figure is a skill published in the GitHub repository XYD-GIS/paper-toolkits (11 stars, last pushed 3mo ago), licensed MIT. It adds 215 tokens to every session and 1,994 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
academic-plotting
Generates publication-quality figures for ML papers from research context. Given a paper section or description, extracts system components and relationships to generate architecture diagrams via Gemini. Given experiment results or data, auto-selects chart type and generates data-driven figures via matplotlib/seaborn.…
ieee-figure
Produce publication-grade result figures for IEEE communications papers (JSAC, TWC, TC/TCOM, WCL, CL) — correct column widths, embedded fonts, vector output (PDF/EPS), grayscale- and colorblind-safe palettes, readable fonts, and the comms-standard plot types: BER/outage vs SNR on a log y-axis, achievable/sum rate vs…
awesome-ieee-report
Generates IEEE-format LaTeX technical reports from project requirements and codebase evidence, using academic prose, pseudocode-only explanations, BibTeX references, and IEEEtran formatting.
paper-figure
A chart-review and design assistant that checks figures and lays out factual design options. It covers choices such as chart type, labels, colours, axes, captions, and error bars.
thesis-figures
A Chinese-language thesis figure workflow that advises on whether to make a visual, chooses a drawing method, checks the underlying data, exports formats, records figures, and reviews the resulting PNG.
academic-plotting
Generates publication-quality figures for ML papers from research context. Given a paper section or description, extracts system components and relationships to generate architecture diagrams via Gemini. Given experiment results or data, auto-selects chart type and generates data-driven figures via matplotlib/seaborn.…