paperbanana

Generate publication-quality figures from textual descriptions via a multi-agent pipeline.

42|3|Updated Feb 4, 2026
One-click install
npx skills add https://github.com/PlutoLei/paperbanana-skill --skill paperbanana
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: paperbanana
Source: https://github.com/PlutoLei/paperbanana-skill/tree/main/plugins/paperbanana/skills/paperbanana
Command: npx skills add https://github.com/PlutoLei/paperbanana-skill --skill paperbanana

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Generate publication-quality diagrams, statistical plots, and presentation slides from plain-text descriptions using PaperBanana's multi-agent pipeline.

Core Features & Use Cases

  • Multi-agent pipeline (Retriever → Planner → Stylist → Visualizer → Critic) to convert textual descriptions into polished visuals
  • Supports diagrams, plots, and slide decks from descriptions with configurable styling
  • Suitable for researchers, educators, and authors who need fast, repeatable visuals for papers, talks, and reports

Quick Start

Provide a text description or path to a description to generate publication-quality visuals.

Frequently Asked Questions about paperbanana

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

FAQPage Schema
How do I generate publication-quality figures from plain text?

You can generate publication-quality figures from plain text by providing a textual description to a multi-agent pipeline that automates the conversion into polished diagrams, plots, and presentation slides.

What is the best way to automate diagram generation for academic papers?

Automating diagram generation for academic papers is best handled by a multi-agent pipeline that processes text descriptions through retrieval, planning, styling, and visualization stages to produce repeatable visuals.

Can I create presentation slides from a text description?

Yes, you can create presentation slides from a text description using a multi-agent pipeline that interprets the text and automatically generates configurable, styled slide decks suitable for talks and reports.

Does the multi-agent pipeline support configurable styling for statistical plots?

Yes, the multi-agent pipeline supports configurable styling for statistical plots, allowing researchers and authors to customize visuals generated from text descriptions to meet specific publication standards.

How does the Retriever Planner Stylist Visualizer Critic pipeline work for visual generation?

The pipeline works by sequentially passing text descriptions through specialized agents: retrieving context, planning the layout, applying styles, visualizing the output, and running a critic check for publication-quality results.

Are there timeout safeguards when generating complex diagrams from text?

Yes, the automated figure generation pipeline includes timeout safeguards and CLI controls to manage the processing of complex textual descriptions into visuals without hanging or failing unexpectedly.