paperbanana-skill

Why use it

Key features

Publication-quality default style
Architecture / system diagrams
Training curve + benchmark plots
Outputs SVG + PDF (LaTeX-ready)
Consistent visual grammar across a paper's figures

Live Demo

What it looks like in practice

paperbanana-skill.replay ▶ ready

0/0

Install

Author repository · Claude Skills documentation

Use Cases

Real-world ways to use paperbanana-skill

Draw your model architecture for a conference paper

👤 ML researchers, grad students mid-paper ⏱ ~30 min intermediate

When to use: Deadline in 5 days, you still don't have a clean architecture figure.

Flow

Describe the model

Use paperbanana-skill. Architecture figure: input → tokenizer → transformer stack (6 layers) → classifier head. Use the skill's default paper style.✓ Copied

→ SVG with clean boxes, arrows, consistent typography
Tweak

Label the transformer block internals — attention + FFN.✓ Copied

→ Updated figure with sub-block detail
Export for LaTeX

Export as PDF sized for a two-column paper.✓ Copied

→ PDF with correct aspect ratio

Outcome: A reviewer-worthy figure without a week of Inkscape.

Pitfalls

Figure over-decorated — Prefer restraint; one accent color, black arrows, lots of whitespace

Combine with: filesystem

Turn training logs into a publication-ready plot

👤 ML researchers with wandb/tensorboard exports ⏱ ~20 min beginner

When to use: You have CSVs of training metrics and need a figure for the paper.

Flow

Feed CSV

Use paperbanana-skill. Plot loss and accuracy curves from training.csv in two-panel layout.✓ Copied

→ Two-panel figure with clean axes
Add baselines

Overlay the baseline model's curves (baseline.csv) as dashed.✓ Copied

→ Overlay added with clear legend

Outcome: A plot that fits the paper's aesthetic.

Keep all figures in a paper visually consistent

👤 Authors with 6+ figures drifting in style ⏱ ~45 min intermediate

When to use: You look at the paper and realize each figure has a different font.

Flow

Define the style

Use paperbanana-skill. Create a shared style: 10pt sans-serif, #1f77b4 primary, black stroke 0.8pt.✓ Copied

→ Style file saved
Apply to all figures

Regenerate figures 1-6 using this style.✓ Copied

→ Consistent batch output

Outcome: Figures that look like they belong in the same paper.

Combine with: filesystem

Combinations

Pair with other MCPs for X10 leverage

paperbanana-skill + filesystem

Organize figures in the paper's /figures directory

Save all figures under paper/figures/ with source.svg + out.pdf per figure.✓ Copied

paperbanana-skill + mermaid-skill

Use Mermaid for casual diagrams, PaperBanana for publication quality

Sketch in Mermaid for discussion; produce with PaperBanana for camera-ready.✓ Copied

Tools

What this MCP exposes

Tool	Inputs	When to call
architecture_diagram	description	System figures
plot_curves	data file, metrics	Training/eval metrics
apply_style	style file, figures	Cross-figure consistency

Cost & Limits

What this costs to run

API quota: None
Tokens per call: Moderate
Monetary: Free
Tip: Lock the style once; apply across figures to avoid re-deciding.

Security

Permissions, secrets, blast radius

Credential storage: None

Data egress: None from skill

Troubleshooting

Common errors and fixes

Exported PDF has wrong font in LaTeX

Embed fonts or use standard PDF font families; check paperbanana's export settings

SVG renders differently in browser vs LaTeX

Stick to PDF for LaTeX; SVG for web/slides

Alternatives