Release Notes - Margarita v0.1.0

Release Date: January 17, 2026

🎉 Overview

This is the initial release of Margarita, a lightweight markup language and Python library for writing, composing, and rendering structured LLM prompts. Margarita is designed for prompt engineering workflows where clarity, versioning, and correctness matter.

✨ Features

Core Functionality

• Parser - Parse .marg template files with support for variables, conditionals, loops, and includes
• Renderer - Render templates with context data to produce final prompts
• Composer - Compose and nest multiple template files together
• CLI Tool - Command-line interface for rendering templates from the terminal

📦 Installation

pip install margarita==0.1.0

🚀 Quick Start

Create a template file hello.marg:

Hello, {{name}}!
Welcome to Margarita templating.

Create a context file context.json:

{
    "name": "World"
}

Render the template:

margarita render hello.marg

🔧 Requirements

• Python >= 3.10
• Dependencies:
  • click >= 8.0.0
  • loguru >= 0.7.3

📚 Documentation

Full documentation is available with examples for:

• Getting Started
• Basic templating
• Conditionals
• Loops
• Include files
• Contexts
• Metadata
• Using with AI agents

📄 License

MIT License - see LICENSE file for details

👥 Author

Kyle Reczek ([email protected])

🔗 Links

• PyPI: https://pypi.org/project/margarita/
• GitHub: https://github.com/Banyango/margarita
• Documentation: https://banyango.github.io/margarita/

📝 Known Limitations

• Working on doing a binary release of the CLI tool. Look for it soon!

🙏 Acknowledgments

Thank you to all early adopters and people who try this out!

Full Changelog: https://github.com/Banyango/margarita/blob/main/CHANGELOG.md