tingxiyu/academic-figure-skill — explained in plain English
Analysis updated 2026-05-18
Turn a CSV of experiment results into a Nature-style volcano plot automatically.
Generate a multi-panel figure from research data with journal-compliant fonts and column widths.
Get a written statistics report alongside a generated figure for traceability.
Reuse a consistent visual style across many figures in the same paper.
| tingxiyu/academic-figure-skill | alirezarezvani/claude-cto-team | gao-ruilin/autorun | |
|---|---|---|---|
| Stars | 100 | 101 | 101 |
| Language | Python | Python | Python |
| Last pushed | — | 2025-12-18 | — |
| Maintenance | — | Quiet | — |
| Setup difficulty | easy | easy | easy |
| Complexity | 3/5 | 2/5 | 2/5 |
| Audience | researcher | pm founder | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires Claude Code or Codex plus a working Python or R environment for rendering.
Academic Figure Skill is a skill package for AI coding assistants such as Claude Code and Codex that automates the process of turning research data into publication-ready scientific figures. Instead of leaving a person to fight with chart formatting by hand, it walks the AI through an eight step process, starting with clarifying what scientific question the figure needs to answer and ending with a finished vector PDF, a 300 dpi PNG preview, and a written statistics report. The skill encodes the visual standards used by top journals such as Nature, Cell, and Science, including approved fonts, fixed column widths, and vector export requirements, along with the visual parameters for 29 common chart types like heatmaps, volcano plots, PCA plots, violin plots, and Sankey diagrams. When a user supplies data and a question, the AI classifies which of four layout patterns fits, proposes a panel plan, checks whether Python or R is available, applies a consistent font and color scheme, and then looks for a matching production script in the assets folder. If one exists it is run as is, and if not, the skill borrows visual settings from a similar chart type. Every generated figure goes through a four pass quality check covering common mistakes, code level compliance, visual and data logic, and the final rendered output, with more than thirty individual checks. Panels are also validated against basic data rules before rendering, such as requiring enough significant genes in a volcano plot, and each figure comes with a report stating sample size, the statistic used, the type of test, and how the source data can be traced. Academic Figure Skill does not replace Python or R plotting itself. It provides structured rules and prior knowledge so the AI's generated plotting code follows journal visual standards, cutting down the manual work of adjusting layout, color, and export settings. It installs by copying the repository into a Claude Code or Codex skills folder and is released under the Apache 2.0 license.
An AI assistant skill that turns research data into publication ready scientific figures that follow top journal formatting standards.
Mainly Python. The stack also includes Python, R.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice and state any changes made.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Verify against the repo before relying on details.