GitHub - kanthorlabs/runit: Run arbitrary python script you have, no setup, no configuration, just run it

🚀 runit

Run Python scripts instantly in containers - no setup, no fuss, just run it.

✨ Features

• 🐍 Zero Python Setup - No local Python installation needed
• 🔒 Isolated Environment - Each script runs in its own container
• 📦 Auto Dependencies - Automatically detects and installs required packages
• 🎯 Simple Interface - Single command to run any Python script
• 🔌 Port Forwarding - Easy exposure of network services
• 🎨 Version Flexible - Use any Python version you need

🚀 Quick Start

Requirements

• Docker: Container runtime (version 20.10 or higher recommended)
• Go: Required for installation from source (version 1.19 or higher)
• Internet connection: For pulling Docker images and installing dependencies

Installation

go install github.com/kanthorlabs/runit/cmd/runit@latest

Basic Usage

# Run a simple script
runit script.py

# Run with specific Python version
runit --platform-version python:3.11-slim script.py

# Expose ports
runit --ports 8080 webapp.py

# Pass arguments to your script
runit github-analyzer.py --arguments="kanthorlabs/runit"

🎯 Command Reference

Syntax

Flags

🎭 Advanced Examples

Note: You can find the source code for these examples in the examples/python directory.

# Web application with multiple ports
runit webapp.py --ports 8080 --ports 9090

# GitHub analyzer with authentication
runit examples/python/github-analyzer.py \
  --arguments="kanthorlabs/runit" \
  --params="--token=xxx"

# Machine learning script with specific Python version
runit examples/python/analyze-image.py \
  --platform-version python:3.11-slim \
  --arguments="https://i.ibb.co/KxcPH8LF/Screenshot-2025-03-11-at-11-09-33.png" \
  --params="--api-key=sk-proj-xxx"

🏗️ Architecture

flowchart TD;
    User -- "runit script.py [flags]" --> CLI[runit CLI]

    subgraph AP [Analysis & Packaging]
        direction TB
        CLI -->|Reads| Script[Script File]
        Script -->|Analyzes Dependencies| Lockfile(requirements.txt)
        CLI -->|Generates| Dockerfile(Dockerfile)
        Script & Lockfile & Dockerfile --> Ctx[Build Context]
    end

    subgraph DO [Docker Operations]
        direction TB
        Ctx -->|Builds Image| Image[Docker Image]
        Image -->|Runs Container| Container[Docker Container]
        Container -->|Streams Logs| Output[STDOUT/STDERR]
        Container -->|Cleanup| Removed[Container Removed]
    end

    AP --> DO
    DO --> Output

🧩 How It Works

1. Analysis: Scans your Python script for dependencies
2. Packaging: Creates a container with your script and dependencies
3. Execution: Runs the script in an isolated environment
4. Cleanup: Automatically removes the container after execution

💡 Pro Tips

• Use python:*-slim images for faster downloads
• Leverage port forwarding for web applications
• Include requirements.txt for explicit dependencies
• Use environment variables for sensitive data

🤝 Contributing

We welcome contributions! Here's how you can help:

• 🐛 Report bugs
• 💡 Suggest features
• 🔧 Submit pull requests
• 📖 Improve documentation

📝 License

MIT License - View License