BunBook

A VS Code notebook extension for running TypeScript with Bun. Fast startup, shared state across cells, inline Plotly charts, and full TypeScript intellisense.

Also comes with a CLI to run and export without vscode (see below).

Install from Marketplace

Features

• Bun runtime — cells execute via Bun with near-instant startup
• Shared state — variables defined in one cell are available in subsequent cells
• Imports — import from local files or node_modules dependencies
• Rich outputs — display() API for HTML, Markdown, JSON, SVG, images, and custom MIME types
• Plotly charts — render interactive charts inline with display.plotly()
• TypeScript intellisense — autocomplete, hover info, and diagnostics across cells
• Formatting — auto-format cells with VS Code's format command
• Kernel restart — reset all state with the restart button in the toolbar
• Standard .ipynb format — notebooks are standard Jupyter files

Requirements

• Bun installed and available in your PATH
• VS Code 1.85.0+

Getting Started

1. Create a .ipynb file
2. Select the TypeScript (Bun) kernel
3. Write TypeScript in code cells
4. Run cells with Shift+Enter or the play button

Output Persistence Toggle

Use the toolbar toggle to control whether cell outputs are saved to the file:

• Outputs saved (foo.ipynb) — outputs are persisted in the file. Use this when you want GitHub or nbviewer to render outputs inline.
• Outputs not saved (foo.no-output.ipynb) — outputs are never written to the file. This keeps git diffs clean since only your code is stored.

Toggling renames the file between .ipynb and .no-output.ipynb. Your unsaved edits are preserved.

Rich Outputs with display()

Use display() to render rich content in cell outputs:

display.html("<h1>Hello World</h1>");
display.markdown("## Bold **text**");
display.json({ name: "test", value: 42 });
display.svg(
  '<svg width="100" height="100"><circle cx="50" cy="50" r="40" fill="red"/></svg>'
);
display.image(pngBuffer); // Buffer | Uint8Array, defaults to image/png
display.image(jpegBuffer, "image/jpeg"); // explicit MIME
display.plotly([{ x: [1, 2], y: [1, 4], type: "scatter" }]); // interactive Plotly chart
display("raw string", "text/plain"); // generic: data + MIME
display({ "text/html": "<b>hi</b>", "text/plain": "hi" }); // multi-MIME

Combine with CLI tools for even more output types:

// Matplotlib plot
const png = await Bun.$`python3 -c ${matplotlibScript}`.arrayBuffer();
display.image(new Uint8Array(png));

// Mermaid diagram
const svg =
  await Bun.$`bunx -p @mermaid-js/mermaid-cli mmdc -i - -o - -e svg < ${diagram}`.text();
display.svg(svg);

Plotly Charts

Use display.plotly() to render interactive charts:

display.plotly([{ x: [1, 2, 3], y: [1, 4, 9], type: "scatter" }], {
  title: "My Chart",
});

The API matches Plotly.js but without the first element parameter:

display.plotly(data, layout?, config?)

Plotly.newPlot(data, layout?, config?) still works as an alias.

Using Dependencies

Create a package.json in your notebook's directory and install packages with bun install. Imports work as expected:

import { mean, standardDeviation } from "simple-statistics";

const data = [1, 2, 3, 4, 5];
console.log(mean(data));

Toolbar & Command Palette

The notebook toolbar ... overflow menu exposes:

All of these are also available via the Command Palette (Cmd+Shift+P / Ctrl+Shift+P).

CLI

The bunbook-cli package lets you run and convert notebooks from the terminal or CI pipelines — no install needed with bunx:

bunx bunbook-cli run notebook.ipynb           # execute cells, save outputs
bunx bunbook-cli export-html notebook.ipynb --run -o report.html
bunx bunbook-cli remove-outputs notebook.ipynb  # strip outputs before committing

See the CLI README for full documentation.

Examples

• hello-world.ipynb — minimal notebook, no setup needed
• plots.ipynb — Plotly charts with shared state across cells
• with-dependencies.ipynb — npm dependencies, local file imports, and linear regression with Plotly
• display.ipynb — rich outputs: HTML, Markdown, JSON, SVG, images, matplotlib, Mermaid, and Vega-Lite

Development

bun install
bun start

This builds, packages, and installs the extension locally. Reload VS Code to pick up changes.