vibefoundry 0.2.20

A local IDE for data science workflows with script running and metadata generation

VibeFoundry IDE

A local desktop IDE for data analysis.

Features

• Local File Management - Browse and manage your project files
• Script Runner - Run Python scripts locally with auto-preview of outputs
• Data Preview - View CSV, Excel, and image files directly in the IDE
• Auto-generated metadata - Per-folder schema digest for input/output data

Installation

# Install the package
pip install -e .

# Build the frontend
cd frontend && npm install && npm run build && cd ..

Usage

# Launch the IDE
vibefoundry

# Or specify a project folder
vibefoundry /path/to/project

Development

# Run frontend dev server
cd frontend && npm run dev

# Run backend separately
python -m vibefoundry.server

Publishing

The publish flow reads the PyPI token from .env (gitignored):

PYPI_TOKEN=pypi-...

.env is expected to be present in the project root — publish.sh sources it automatically. Bump the version and ship:

./publish.sh 0.1.307

Architecture

• frontend/ - React-based UI (Vite + React)
• src/vibefoundry/ - Python backend (FastAPI)
  • server.py - Main API server
  • watcher.py - File change detection
  • runner.py - Script execution
  • metadata.py - Metadata generation