PaperBanana
Turn paper text into publication-style diagrams and statistical plots through a multi-agent MCP pipeline instead of manual figure tooling.
Overview
PaperBanana is an MCP server for the Build phase that generates academic diagrams and statistical plots from text using multi-agent AI.
What is this MCP server?
- Generates academic diagrams from natural-language descriptions
- Produces statistical plots from text-specified data stories
- Multi-agent AI orchestration behind a single MCP surface
- PyPI package paperbanana with stdio MCP transport
- Version 0.1.1 PaperBanana server from llmsresearch
- PyPI identifier paperbanana at version 0.1.1
- Stdio transport MCP server schema PaperBanana
- Multi-agent AI pipeline described as core capability in package metadata
Community signal: 2k GitHub stars.
What problem does it solve?
Solo builders waste hours redrawing architecture sketches and stats figures that should stay in sync with evolving docs.
Who is it for?
Indie researchers and technical writers who want agent-generated figures during doc and content builds.
Skip if: Production dashboard BI, brand-polished marketing design systems, or teams needing audited financial charts without human review.
What do I get? / Deliverables
Your agent invokes PaperBanana MCP tools to draft diagrams and plots you can drop into papers, repos, or long-form docs.
- Generated academic-style diagrams from text prompts
- Statistical plot outputs derived from agent-authored specifications
- Figures ready to embed in docs or manuscripts after human review
Recommended MCP Servers
Journey fit
Figure production belongs while you are building artifacts—README visuals, technical docs, or research outputs—not during raw ideation. Docs is the shelf for diagrams and plots that explain systems, methods, or results alongside written material.
How it compares
Figure-generation MCP using multi-agent LLM flows, not a general Figma or matplotlib cheat-sheet skill.
Common Questions / FAQ
Who is PaperBanana for?
Builders and researchers who publish technical writing and want MCP-driven academic diagrams and statistical plots.
When should I use PaperBanana?
While writing or revising docs in the build phase when you need figures that match explanatory text.
How do I add PaperBanana to my agent?
Install the paperbanana package from PyPI, configure stdio MCP in Claude Code or Cursor, and ensure your environment supplies whatever API keys the multi-agent backend expects.