Back to search
CCF-Figure
SkillGitHub
CCF-Figure —— 为 AI / 计算机科学论文自动生成顶会级科研配图的智能助手。
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Checked today
Badge for your README
Markdown
[](/listings/deepshare-official--ccf-figure)HTML
<a href="/listings/deepshare-official--ccf-figure"><img src="/api/listings/deepshare-official--ccf-figure/badge.svg" alt="findsafeskills safety rating" /></a>Install
i
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Claude Code
# Option 1: Install as a user-level Claude Code skill, available across projects git clone https://github.com/Deepshare-Official/CCF-Figure \ ~/.claude/skills/ccf-figure # Option 2: Install as a project-level Claude Code skill, available only in the current project git clone https://github.com/Deepshare-Official/CCF-Figure \ .claude/skills/ccf-figure
Codex
# Option 1: Install as a user-level Codex skill, available across projects git clone https://github.com/Deepshare-Official/CCF-Figure \ ~/.agents/skills/ccf-figure # Option 2: Install as a project-level Codex skill, available only in the current project git clone https://github.com/Deepshare-Official/CCF-Figure \ .agents/skills/ccf-figure
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
SKILL.md · Manifest · 165 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- ---
- name: scientific-figure-generator
- version: "2.0"
- description: >
- Generate publication-ready scientific figures for AI/CS research papers.
- Automatically classifies paper type, selects the optimal diagram structure,
- and produces figures meeting the visual standards of NeurIPS, ICML, CVPR,
- Nature Machine Intelligence, and other top venues.
- platforms:
- - claude
- - codex
- author: Deepshare / Deepshare
- source: https://mp.weixin.qq.com/s/qPqfvBILEu5ascyZe6guAw
- ---
- # Scientific Figure Generator — Skill Instructions
- ## Role
- You are an expert scientific illustrator specializing in computer science and AI research.
- Your mission is **not** to simply beautify paper content, but to transform complex scientific
- text into logically accurate, structurally clear, information-dense, low-noise figures that
- meet the visual standards of top-tier conferences and journals.
- You must first understand the paper's research type, core problem, method mechanism,
- experimental logic, and key contributions — **then** decide on the most appropriate diagram
- structure. You are **prohibited** from mechanically applying the "left-input → center-model →
- right-output" three-column template unless the paper genuinely fits that structure.
- ---
- ## SOP — Standard Operating Procedure
- ### Step 1 · Input Validation
- Ask the user to provide one or more of the following (the more the better):
- - Paper title
- - Abstract
- - Method / approach section (key paragraphs)
- - Core contributions (bullet points)
- - Any specific visual requirement (e.g., "focus on the feedback loop", "show the three modules side-by-side")
- If the user provides **only an abstract**, warn them:
- > "Abstracts describe results, not structure. For a better figure, please also paste the method section or describe the key modules and their relationships."
- ---
- ### Step 2 · Paper Type Classification
- Classify the paper into **one primary type** (and optionally a secondary type):
- | ID | Type | Trigger Keywords / Signals |
- |----|------|---------------------------|
- | A | **Method Paper** | proposes a new model/framework/algorithm/loss/training strategy |
- | B | **Mechanism / Analysis Paper** | ablation study, module analysis, probing, attention visualization |
- | C | **Benchmark / Evaluation Paper** | new dataset, evaluation suite, leaderboard, task taxonomy |
- | D | **Scaling Law / Empirical Analysis** | scaling curves, compute budget, performance trend, law fitting |
- | E | **Robot / Embodied AI Paper** | perception-action loop, manipulation, navigation, sim-to-real |
- | F | **Interdisciplinary / AI for X** | medical AI, AI for science, smart manufacturing, cybersecurity |
- | G | **Survey / Position Paper** | literature taxonomy, research roadmap, comparison matrix |
- > See `references/figure-types.md` for the full type × structure mapping table.
- ---
- ### Step 3 · Diagram Structure Selection
- Based on the paper type, **automatically select** the most appropriate diagram structure:
- | Paper Type | Recommended Structure(s) |
- |-----------|--------------------------|
- | A — Method | Method overview, Model architecture, Layered pipeline |
- | B — Mechanism | Local zoom-in diagram, Comparison diagram, Ablation matrix |
- | C — Benchmark | Evaluation flow, Multi-panel comparison, Data construction pipeline |
- | D — Scaling | Experimental matrix, Trend curves panel, Variable-relationship diagram |
- | E — Robot/Embodied | **Closed-loop feedback diagram** (perception → language → planning → action → environment) |
- | F — Interdisciplinary | Cross-domain application framework (domain data + AI model + task + metrics) |
- | G — Survey | Taxonomy tree, Timeline, Comparison matrix |
- ---
- ### Step 4 · Generate the Full Image Prompt
- Assemble the final prompt by combining:
- 1. The **base visual specification** (from `references/prompt-template.md`, Section A)
- 2. The **paper-type-specific instruction** (from `references/prompt-template.md`, Section B)
- 3. The **user's paper content** (pasted verbatim at the end)
- Then pass this assembled prompt to the image generation model.
- **For Claude**: Use the built-in image generation capability directly.
- **For Codex**: Output the assembled prompt and instruct the user to paste it into
- ChatGPT (GPT Image 2 / DALL·E 3) or another image generation API.
- ---
- ### Step 5 · Self-Check Before Delivery
- Before presenting the result, verify against the **Five Common Failure Modes**
- (see `references/figure-types.md`, Section 3):
- - [ ] Did I use only the abstract? (if yes → request method details)
- - [ ] Did I request vague style words like "high-end" or "sci-fi"? (if yes → replace with specific venue style)
- - [ ] Did I tell the model the paper type? (if no → re-run Step 2)
- - [ ] Is the information density excessive? (if yes → trim to one clear visual storyline)
- - [ ] Is there excessive decoration? (if yes → enforce flat vector rules)
- ---
- ### Step 6 · Iteration Protocol
- If the user requests revisions:
- 1. Ask which specific element to change (structure, color, text, emphasis)
- 2. Diagnose whether the change is **structural** (re-run Step 3) or **cosmetic** (tweak prompt)
- 3. Generate a revised prompt and re-run generation
- 4. Maximum 3 revision rounds before suggesting a fundamentally different diagram type
- ---
- ## Quick-Start Template
- When a user wants to generate a figure immediately, respond with:
- ```
- Please provide the following for your paper:
- 1. **Title** (required)
- 2. **Abstract** (required)
- 3. **Method section** or key module descriptions (strongly recommended)
- 4. **Main contributions** (optional but helpful)
- 5. **Any specific visual preference** (optional — e.g., "emphasize the two-stage pipeline", "use a closed-loop structure")
- I will automatically classify your paper type, select the best diagram structure,
- and generate a top-conference-quality figure prompt for you.
- ```
- ---
- ## Visual Specification Summary
- | Property | Rule |
- |----------|------|
- | Background | Pure white (`#FFFFFF`) or very light gray (`#F5F5F5`) |
- | Style | 2D flat vector — NO 3D perspective, NO neon gradients, NO decorative textures |
- | Color usage | Functional only (encode logic hierarchy, not decoration) |
- | Color palette | Max 3 color groups; large areas = low-saturation cool gray / light blue / off-white |
- | Accent color | Sparse use of orange, cyan-green, or deep blue for key innovations |
- | Typography | Horizontal only; max 2 font sizes; Chinese for labels, English for proper nouns |
- | Symbols | Data/input = parallelogram or card stack; Model = rectangle; Decision = diamond; Feedback = closed-loop arrow |
- | Target venues | NeurIPS, ICML, ICLR, CVPR, ICCV, ACL, KDD, SIGIR, AAAI, Nature MI, Science Robotics, IEEE TPAMI |
- ---
- ## File Structure
- ```
- scientific-figure-generator/
- ├── SKILL.md ← this file (main workflow)
- ├── references/
- │ ├── prompt-template.md ← full image generation prompt (copy-paste ready)
- │ └── figure-types.md ← 11 diagram types × paper type table + 5 failure modes
- └── README.md ← GitHub-ready bilingual documentation
- ```
README.md · README · 267 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- # CCF Figure
- <p align="center">
- <img src="https://img.shields.io/badge/Platform-Claude%20%7C%20Codex-blue?style=flat-square" />
- <img src="https://img.shields.io/badge/Version-2.0-brightgreen?style=flat-square" />
- <img src="https://img.shields.io/badge/Source-DeepShare%20深度之眼-orange?style=flat-square" />
- <img src="https://img.shields.io/badge/License-MIT-lightgrey?style=flat-square" />
- </p>
- <p align="center">
- A skill for generating publication-ready scientific figures from AI/CS research papers.<br>
- Classify paper type first, then select the optimal diagram structure automatically.
- </p>
- ---
- <!-- Chinese version (default) -->
- ## 中文说明
- <details open>
- <summary>点击展开 / 收起</summary>
- ### 是什么
- `CCF Figure` 是一个专门为 AI / 计算机科学领域的研究人员设计,帮助你将论文内容转化为**顶会顶刊级科研配图**的工具。
- 核心理念来自「深度之眼」2026年6月11日发布的文章《科研绘图提示词第二弹》:
- **先判断论文类型,再选择最合适的图示结构**,而不是千篇一律地套用“左输入-中模型-右输出”三段式模板。
- ### 支持平台
- | 平台 | 使用方式 |
- |-----|---------|
- | **Claude Code** | 安装到 skills 目录后,在项目中自动调用 |
- | **Codex** | 安装到 skills 目录后,生成完整提示词并辅助调用图像模型 |
- ### 功能特性
- - **自动分类**:识别 7 种论文类型(方法 / 机制 / 评测 / 规律 / 机器人 / 交叉 / 综述)
- - **结构自选**:从 11 种图示结构中自动匹配最优布局
- - **完整提示词**:内置中英文双语提示词模板,可直接复制使用
- - **五大翻车预防**:内置 Self-Check 清单,避免 AI 绘图的常见错误
- - **迭代协议**:标准化修改流程,最多 3 轮迭代
- ### 支持的论文类型 × 图示结构
- | 论文类型 | 推荐图示结构 |
- |--------|------------|
- | A · 方法类 | 方法总览图 / 模型架构图 |
- | B · 机制 / 分析类 | 机制示意图(局部放大)/ 对比式方法图 |
- | C · 评测 / 基准类 | 评测流程图 / 数据构建流程图 |
- | D · 规律发现类 | 实验规律总结图 / 实验矩阵图 |
- | E · 机器人 / 具身智能 | **闭环反馈图**(感知 → 规划 → 执行 → 反馈) |
- | F · 交叉方向 | 跨学科应用框架图 |
- | G · 综述 / 位置 | 分类树 / 时间线 / 对比矩阵 |
- ### 视觉规范摘要
- | 属性 | 规则 |
- |-----|------|
- | 背景 | 纯白 `#FFFFFF` 或极浅灰 `#F5F5F5` |
- | 风格 | 二维扁平化矢量 — 禁止 3D 透视 / 霓虹渐变 / 装饰纹理 |
- | 配色 | 功能性配色;最多 3 个色彩群组;强调色限橙 / 青绿 / 深蓝 |
- | 文字 | 水平排布;最多 2 种字号;中文标签 + 英文专有名词 |
- | 目标刊会 | NeurIPS, ICML, ICLR, CVPR, ACL, Nature MI, IEEE TPAMI 等 |
- ### 安装方式
- #### Claude Code
- ```bash
- # 方式一:安装到 Claude Code 个人级 skills 目录(推荐,所有项目通用)
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- ~/.claude/skills/ccf-figure
- # 方式二:安装到当前项目的 Claude Code skills 目录(仅当前项目可用)
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- .claude/skills/ccf-figure
- ```
- #### Codex
- ```bash
- # 方式一:安装到 Codex 个人级 skills 目录(推荐,所有项目通用)
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- ~/.agents/skills/ccf-figure
- # 方式二:安装到当前项目的 Codex skills 目录(仅当前项目可用)
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- .agents/skills/ccf-figure
- ```
- ### 使用方式
- 安装完成后,在 Claude Code 或 Codex 中直接说:
- ```text
- 帮我为这篇论文生成科研配图:
- [发送你的论文原文,或粘贴论文标题 + 摘要 + 方法部分]
- ```
- `CCF Figure` 会自动:
- 1. 判断论文类型
- 2. 选择最合适的图示结构
- 3. 组装完整提示词
- 4. 生成图像或输出可直接用于图像模型的完整提示词
- ### 文件结构
- ```text
- CCF-Figure/
- ├── SKILL.md ← 主工作流(Skill 入口)
- ├── references/
- │ ├── prompt-template.md ← 完整提示词模板(中英双语,可直接复制)
- │ └── figure-types.md ← 11 种图示类型规范 + 五大翻车问题
- └── README.md ← 本文件
- ```
- ### 建议的 SKILL.md 头部配置
- 为了让 Claude Code 和 Codex 更稳定识别,建议 `SKILL.md` 开头使用如下 metadata:
- ```markdown
- ---
- name: ccf-figure
- description: Generate publication-ready scientific figures for AI and computer science papers. Use when the user provides a paper title, abstract, method section, or full paper and wants a top-conference-style research diagram.
- ---
- # CCF Figure
- ```
- </details>
- ---
- <!-- English version -->
- ## English Documentation
- <details>
- <summary>Click to expand / collapse</summary>
- ### What It Is
- `CCF Figure` is a Skill for researchers in AI and computer science.
- It transforms paper content into **publication-quality scientific figures** that meet the visual standards of top conferences and journals.
- The core methodology comes from DeepShare’s June 2026 article:
- **Classify the paper type first, then select the optimal diagram structure** — rather than mechanically applying a generic “left-input → center-model → right-output” template.
- ### Supported Platforms
- | Platform | How it works |
- |----------|-------------|
- | **Claude Code** | Install it into the skills directory, then invoke it naturally in your project |
- | **Codex** | Install it into the skills directory, then use it to assemble complete scientific-figure prompts |
- ### Features
- - **Auto-classification**: Identifies 7 paper types — method, mechanism, benchmark, scaling, robotics, interdisciplinary, and survey
- - **Structure auto-selection**: Automatically matches the best layout from 11 diagram structures
- - **Complete prompt library**: Bilingual Chinese / English prompt templates, ready to copy and use
- - **5-failure-mode prevention**: Built-in self-check checklist to avoid common AI figure mistakes
- - **Iteration protocol**: Standardized revision workflow, up to 3 rounds
- ### Paper Type × Diagram Structure
- | Paper Type | Recommended Diagram |
- |-----------|---------------------|
- | A · Method | Method Overview / Model Architecture |
- | B · Mechanism / Analysis | Mechanism Zoom-in / Contrastive Diagram |
- | C · Benchmark / Evaluation | Evaluation Pipeline / Data Construction Pipeline |
- | D · Scaling Law / Empirical | Trend Panel / Experimental Matrix |
- | E · Robot / Embodied AI | **Closed-Loop Feedback Diagram** |
- | F · Interdisciplinary / AI for X | Cross-domain Application Framework |
- | G · Survey / Position | Taxonomy Tree / Timeline / Comparison Matrix |
- ### Visual Specification
- | Property | Rule |
- |----------|------|
- | Background | Pure white `#FFFFFF` or very light gray `#F5F5F5` |
- | Style | 2D flat vector — NO 3D, NO neon gradients, NO decorative textures |
- | Color | Functional only; max 3 color groups; accent colors limited to orange, cyan-green, and deep blue |
- | Typography | Horizontal only; max 2 font sizes; Chinese labels + English proper nouns |
- | Target venues | NeurIPS, ICML, ICLR, CVPR, ACL, Nature MI, IEEE TPAMI, etc. |
- ### Installation
- #### Claude Code
- ```bash
- # Option 1: Install as a user-level Claude Code skill, available across projects
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- ~/.claude/skills/ccf-figure
- # Option 2: Install as a project-level Claude Code skill, available only in the current project
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- .claude/skills/ccf-figure
- ```
- #### Codex
- ```bash
- # Option 1: Install as a user-level Codex skill, available across projects
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- ~/.agents/skills/ccf-figure
- # Option 2: Install as a project-level Codex skill, available only in the current project
- git clone https://github.com/Deepshare-Official/CCF-Figure \
- .agents/skills/ccf-figure
- ```
- ### Usage
- After installation, simply say in Claude Code or Codex:
- ```text
- Generate a scientific figure for this paper:
- [Paste your paper, or paste the paper title + abstract + method section]
- ```
- `CCF Figure` will automatically:
- 1. Classify the paper type
- 2. Select the most appropriate diagram structure
- 3. Assemble the complete image-generation prompt
- 4. Generate the figure or output a complete prompt ready for an image model
- ### File Structure
- ```text
- CCF-Figure/
- ├── SKILL.md ← Main workflow / Skill entry point
- ├── references/
- │ ├── prompt-template.md ← Full prompt template, bilingual and copy-paste ready
- │ └── figure-types.md ← 11 diagram type specs + 5 failure modes
- └── README.md ← This file
- ```
- ### Recommended SKILL.md Metadata
- For better compatibility with Claude Code and Codex, it is recommended to start `SKILL.md` with the following metadata:
- ```markdown
- ---
- name: ccf-figure
- description: Generate publication-ready scientific figures for AI and computer science papers. Use when the user provides a paper title, abstract, method section, or full paper and wants a top-conference-style research diagram.
- ---
- # CCF Figure
- ```
- </details>
- ---
- ## License
- MIT License — feel free to use, modify, and distribute with attribution.
- ---
- <p align="center">
- Made with ❤️ for the AI research community · 深度之眼 Deepshare
- </p>
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.