Verified:GitHub

paper-figure

Generate publication-ready figures and LaTeX tables from experiment results.

Updated Apr 21, 2026

Search Tags:#data visualization #reproducibility #experiment analysis #latex #figures #paper writing #scientific plotting

Ready to run

npx skills add https://github.com/Shallow-W/llm-wiki --skill paper-figure-shallow-w

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: paper-figure
Source: https://github.com/Shallow-W/llm-wiki/tree/main/.claude/skills/paper-figure
Command: npx skills add https://github.com/Shallow-W/llm-wiki --skill paper-figure-shallow-w

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the time-consuming manual work of turning experiment results into clean, publication-ready figures and comparison tables for a paper.

Core Features & Use Cases

Data-driven figure generation: Produces common ML plots such as line charts, bar charts, scatter plots, heatmaps, and box/violin plots from experiment data.
Paper comparison outputs: Generates multi-panel figures and LaTeX-ready comparison tables (including ablations and method comparisons).
Reproducible workflow and consistent styling: Creates per-figure generation scripts and ensures consistent fonts, sizes, and print-friendly rules (e.g., vector PDF output, no titles inside figures).

Quick Start

Ask the paper-figure skill to generate the needed paper figures from your figure plan and experiment data by running: generate the paper figures using PAPER_PLAN.md and the available JSON/CSV data.

Frequently Asked Questions about paper-figure

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate publication-ready figures from CSV data for my paper?▼

To generate publication-ready figures, provide your CSV data and PAPER_PLAN.md file. The skill creates per-figure scripts and outputs vector PDFs with LaTeX include snippets for consistent styling.

How do I create LaTeX comparison tables from experiment results?▼

Creating LaTeX comparison tables from experiment results involves reading JSON or screen logs to structure ablations and method comparisons. The process produces LaTeX-ready tables that integrate directly into your paper manuscript.

What types of scientific plots can I make for an ML paper?▼

For an ML paper, you can make line charts, bar charts, scatter plots, heatmaps, and box or violin plots. These scientific plots support multi-panel layouts and apply consistent publication styling defaults.

Can I use JSON log files to plot training curves automatically?▼

Yes, you can use JSON log files to plot training curves automatically. The workflow reads JSON inputs alongside PAPER_PLAN.md to produce reproducible per-figure scripts and vector PDF outputs without manual plotting.

Does the generated figure styling follow academic publication standards?▼

The generated figure styling follows academic publication standards by enforcing consistent fonts, sizes, and print-friendly rules. It outputs vector PDFs and omits titles inside figures to meet professional formatting expectations.

How do I ensure reproducibility when generating multi-panel figures?▼

To ensure reproducibility when generating multi-panel figures, the workflow creates dedicated per-figure generation scripts. This allows you to regenerate the exact same vector PDF outputs from your experiment data consistently.