Happy Figure Skill
The skill turns paper text, methods and captions into structured prompts for scientific figures, tuned to the chosen image model.
- 180 ★ on the repository
- Last commit
Minimal access
Why this level
- The skill writes prompt text and sends nothing itself. Article content goes to the agent's model.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/happy-figure-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skill writes prompt text and sends nothing itself. Article content goes to the agent's model.
This is third-party code. Review the repository files before installing.
What it does
Happy Figure Skill is a skill for Claude Code and Codex. The agent reads research material and assembles a prompt for an image model. The skill itself does not generate pictures. Prompts are built along three axes: scientific field, figure type and model traits. Named models include Nano Banana Pro, Nano Banana 2, Qwen Image 2 and GPT Image 2. The `references/` folder holds templates for fields (computer science, materials chemistry, biomedicine, earth science), graphical abstracts, technical roadmaps and mechanism diagrams. A separate style-transfer mode takes layout, colors and line work from a reference picture while the scientific content stays your own. The `scripts/extract_research_doc.py` script pulls text out of a research document.
Who it is for. Graduate students and researchers who prepare figures for a paper or thesis and write image-generator prompts by hand.
Good fit when
- You need a prompt for a paper's graphical abstract.
- A methods section must become a technology roadmap.
- One figure is prepared for several models and the results compared.
- A figure is styled after a journal using someone else's picture as a sample.
Not a fit when
- You need a finished picture straight from the skill.
- The figure holds exact data and plots that a generator cannot be trusted with.
- You want to reproduce someone else's illustration including its scientific content.
Example request
Use the skill to read this abstract and write a graphical-abstract prompt for Nano Banana Pro with short English labels.Limitations
The skill outputs prompt text only, so a separate model must draw the picture. Generators often garble text inside images, so check labels by hand. Documentation is mostly Chinese with a short English version. The last repository commit dates from June 2026. The repository declares no license, so check usage terms before publishing figures made with the skill's material. Models and their abilities change fast, so names in the skill's references may lag behind current versions.
How to disable. Delete the skill folder from the agent's skills directory. Saved prompts remain in your files.
FAQ
Does the skill generate images itself?
No. It prepares prompt text and the chosen model draws the picture: Nano Banana Pro, Qwen Image 2, GPT Image 2 or another.
How do I call the skill?
In Claude Code or Codex write "Use $happy-figure-skill" and paste the material. You can name the figure type, field and target model.
Can I borrow another figure's style?
Yes, but only layout, colors, lines and fonts. The skill does not copy the sample's scientific content.
Which figures work best?
Structural ones: model architectures, technical roadmaps, mechanism diagrams and graphical abstracts. Build exact data plots with a separate tool.
Do I need Chinese?
No. The main description is Chinese, but `README_EN.md` exists and prompts can be requested in English.
Discussion
No questions or reviews yet.
Related
SkillsA set of skills for Claude Code and Codex that assemble detailed image prompts from an element library and composition schemes
Nano Banana Pro Prompts Recommend
AI Image Prompt Recommender
A skill that picks ready image-generation prompts from a large library, shows sample images and assembles a prompt for your text
A CLI and Claude Code skill for Gemini images from 512 to 4K, with reference-based edits and transparent assets
A Claude Code skill that compiles a vague request into a finished gpt-image-2 prompt and checks it with a validator script
A Claude Code skill for generating, editing and reviewing images with Google Gemini models, with a plan and approval before every paid request
Seven Claude Code skills and a prompt library for the Seedance, Kling and HappyHorse video models: prompts, storyboards, debugging
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Happy Figure Skill (this card) | no | 180 | |
| Academic Figure Skill | no | 555 | provides scientific figure presets; Happy Figure routes by field, figure type and model |
| SciPilot Figure | no | 2,789 | works within its own style set; Happy Figure maps material onto several generators |
README in short
It picks a route by three traits: scientific field, figure type and model traits. The same material therefore gets different prompts for different generators.