Skip to content

Latest commit

 

History

History
89 lines (66 loc) · 3.4 KB

language_support.md

File metadata and controls

89 lines (66 loc) · 3.4 KB

Language Support in Lux

Lux is designed to be language-agnostic, allowing you to build agents in your preferred programming language. This guide provides an overview of language support in Lux and how to use different programming languages in your agents.

Currently Supported Languages

Python Integration

Learn more about Python integration

Python is a first-class citizen in Lux, with deep integration into the framework. You can:

  • Write Python code directly in your Elixir files using ~PY sigils
  • Add custom Python modules under priv/python and import them
  • Use any Python package through pip/poetry
  • Leverage Python's rich ML and data science ecosystem

🔥 Coming Soon: Define Prisms, Beams, and other Lux components entirely in Python! This will allow you to write your agents' logic completely in Python while still leveraging Lux's powerful orchestration capabilities.

Node.js/JavaScript Integration

Learn more about Node.js integration

Node.js integration allows you to leverage the vast JavaScript ecosystem:

  • Write JavaScript/TypeScript code using ~JS sigils
  • Use NPM packages in your agents
  • Support for modern ES modules and async/await
  • Full access to Node.js APIs

Future Language Support

Rust Integration (Coming Soon)

Native Rust integration is under development, which will provide:

  • High-performance components for compute-intensive tasks
  • Direct FFI integration for optimal performance
  • Access to Rust's rich ecosystem
  • Memory-safe interop with Elixir

Custom Language Integration

Want to add support for your favorite language? Lux provides a Language Integration Protocol that allows you to:

  • Add new language runtimes
  • Define type conversion rules
  • Handle package management
  • Implement error handling

Check out our Contributing Guide for details on adding language support.

Best Practices

  1. Choose the Right Language

    • Use Python for ML/AI tasks and data processing
    • Use Node.js for web integration and text processing
    • Use Elixir for coordination and state management
    • Consider Rust (coming soon) for performance-critical components

  2. Package Management

    • Add custom Python modules under priv/python/
    • Use poetry for Python dependency management
    • Use npm/yarn for Node.js dependencies
    • Follow the language's best practices for versioning

  3. Performance Considerations

    • Minimize cross-language calls
    • Batch operations when possible
    • Use appropriate data serialization
    • Consider memory usage patterns

  4. Security

    • Validate all inputs
    • Control package versions
    • Follow security best practices for each language
    • Implement proper sandboxing

Examples

Check out these examples of language integration:

Contributing

Want to add support for a new language? Check our Contributing Guide and:

  1. Review the Language Integration Protocol
  2. Create a proof of concept
  3. Add comprehensive tests
  4. Document the integration
  5. Submit a pull request

For detailed examples and up-to-date documentation, visit hexdocs.pm/lux.