Yeastbook

A polyglot notebook powered by Bun.
TypeScript-first with Python support. One command, zero config.

---

Quickstart

bunx @codepawl/yeastbook new

That's it. No Python, no conda, no kernel installs. Opens in your browser in under 1 second.

---

Why Yeastbook?

	Jupyter	Marimo	Yeastbook
Language	Python (+ kernels)	Python	TypeScript + Python
Runtime	IPython + ZeroMQ	Python	Bun (fast, single binary)
Setup	conda/pip + kernel install	pip install	bunx or single binary
Top-level await	No	No	Yes
Type checking	No	No	Monaco + TypeScript
Format	.ipynb (complex JSON)	.py	.ybk + .ipynb + .ybk.md
Package install	%pip install	import	%install lodash
Cross-language	Kernels (no shared state)	No	YeastBridge (shared data)
SQL	Needs extensions	No	%sql built-in

---

Features

Editor Full TypeScript with Monaco, IntelliSense, type checking Multi-cursor support (Alt+Click, Ctrl+D, cross-cell) Cell folding, find & replace across cells (Ctrl+Shift+H) Command palette (Ctrl+Shift+P) Execution Top-level await works out of the box Variables persist across cells Session snapshots survive server restarts (24h) Cell execution queue with cancel

Output Charts (Chart.js), Vega/Vega-Lite visualizations Interactive DataFrame viewer with sort, filter, pagination JSON trees, data tables, HTML rendering LaTeX/KaTeX math rendering Matplotlib & PIL image display (Python cells) Ecosystem VS Code extension with full notebook support Jupyter import/export (.ipynb) Readable diff format (.ybk.md) for clean git diffs Plugin API for custom output renderers

---

Tech Stack

• Runtime: Bun — fast JavaScript/TypeScript runtime
• Language: TypeScript (with Python cell support via persistent daemon)
• Editor: Monaco (VS Code's editor engine) with IntelliSense
• UI: React, bundled via Bun's built-in bundler
• Database: bun:sqlite for SQL cells
• Format: .ybk (JSON-based) with .ipynb import/export
• CI/CD: GitHub Actions — test, build, publish on tag push

---

Install

# Run instantly (no install needed)
bunx @codepawl/yeastbook new

# Or install globally
bun install -g @codepawl/yeastbook
yeastbook new

# Or download a binary (no Bun required)
curl -fsSL https://github.com/codepawl/yeastbook/releases/latest/download/install.sh | bash

# Or via Homebrew (macOS/Linux)
brew install codepawl/tap/yeastbook

VS Code Extension

Search "Yeastbook" in VS Code Extensions, or install from the Marketplace.

System Check

yeastbook doctor    # Verify Bun, Python, venv, ports

---

Python Support

Toggle any code cell to Python with the language badge or press L in command mode.

import numpy as np
data = np.random.randn(1000)
print(f"Mean: {data.mean():.4f}, Std: {data.std():.4f}")

YeastBridge — share data between TypeScript and Python:

// TypeScript cell
yb.push("config", { lr: 0.001, epochs: 10 })

# Python cell
config = yb.get("config")
print(config["lr"])  # 0.001

Yeastbook auto-detects virtualenvs (.venv/ or venv/) and injects them automatically.

---

SQL Support

Query SQLite databases and CSV files directly in your notebook:

%sql import sales.csv as sales
%sql SELECT category, SUM(amount) as total FROM sales GROUP BY category ORDER BY total DESC

%sql attach analytics.db
%sql @analytics SELECT * FROM events WHERE date > '2024-01-01' LIMIT 100

Results render as interactive tables with sorting, filtering, and pagination.

---

Rich Output

// Chart.js
;({ __type: "chart", data: [10, 20, 30], config: { chartType: "bar", title: "Sales" } })

// Vega-Lite
;({ __type: "vega", spec: { $schema: "...", data: { values: [...] }, mark: "bar", encoding: {...} } })

// DataFrame
;({ __type: "dataframe", columns: ["name", "age"], data: [{name: "Alice", age: 30}], shape: [100, 2] })

// LaTeX
;({ __type: "math", latex: "E = mc^2" })

// HTML
;({ __type: "html", html: "<h1 style='color: tomato'>Hello</h1>" })

// Table (any array of objects auto-renders)
[{ name: "Alice", age: 30 }, { name: "Bob", age: 25 }]

---

CLI Reference

yeastbook new                        # Create new .ybk notebook
yeastbook new --ipynb                # Create new .ipynb notebook
yeastbook new --template <name>      # Create from template
yeastbook <file.ybk>                 # Open existing notebook
yeastbook export <file.ybk>          # Convert .ybk -> .ipynb
yeastbook import <file.ipynb>        # Convert .ipynb -> .ybk
yeastbook export-md <file.ybk>       # Convert .ybk -> .ybk.md (readable diff)
yeastbook import-md <file.ybk.md>    # Convert .ybk.md -> .ybk
yeastbook export-script <file.ybk>   # Export to .ts script
yeastbook strip-outputs <file.ybk>   # Remove outputs for clean commits
yeastbook doctor                     # Check system requirements

---

Keyboard Shortcuts

Shortcut	Action
Shift+Enter	Run cell & advance
Ctrl+Enter	Run cell & stay
Ctrl+S	Save
Ctrl+Shift+P	Command palette
Ctrl+Shift+E	Toggle presentation mode
Ctrl+Shift+H	Find & replace across cells
Ctrl+Shift+D	Select across all cells
A / B	Add cell above / below (command mode)
D D	Delete cell (command mode)
M / Y	Switch to markdown / code
L	Toggle TypeScript / Python

---

Development

git clone https://github.com/codepawl/yeastbook
cd yeastbook
bun install
bun run dev          # Build UI + start dev server with hot reload
bun test             # Run tests (285 tests)
bun run build:all    # Full build (UI + embed + all platform binaries)

Project Structure

packages/
  core/    @codepawl/yeastbook-core   — shared types, transforms, format conversion
  app/     @codepawl/yeastbook        — CLI, server, kernel, SQL engine
  ui/      @codepawl/yeastbook-ui     — React UI (Monaco editor, rich outputs)
  vscode/  vscode-yeastbook           — VS Code extension (notebook API)

Publishing

Releases are automated via GitHub Actions on tag push (v*.*.*). Manual publish:

npm login
bun run publish:all

---

Contributing

See CONTRIBUTING.md. PRs welcome!

License

MIT — Made by CodePawl

Built with Claude Code as coding assistant.