margarita 0.6.2

A lightweight markup language and Python library for writing, composing, and rendering structured LLM prompts.

MARGARITA

www.margarita.run

Margarita aims to make writing Agents as easy as writing Markdown.

It provides two file formats:

• .mg — A templating language that renders dynamic prompts to Markdown. Extends Markdown with variables, conditionals, loops, and includes.
• .mgx — An agent scripting language that extends .mg with agentic execution: state, memory, tool calls, user input, and more.

Features

• Agentic execution — run .mgx scripts as stateful agents with memory and tool calls in a TUI.
• Composable — .mg files can be split, reused, and nested with [[ include.mg ]] syntax.
• Logical structures — conditionals and loops for dynamic prompt generation. if, else, elif, and for blocks supported.
• Context management — manage agent context with @effect context.
• Memory — persist variables across runs with @memory.
• Input — prompt the user for input during a run with @effect input.
• Tools — register Python functions as LLM-callable tools with @effect tools.
• Function calls — execute Python functions directly and save their result to state with @effect func.
• Sub Agents — call other .mgx files as sub-agents with @effect exec.
• Metadata — attach version and description metadata alongside your prompts. parameters field for defining expected context variables.

Requirements

• A Github Copilot subscription or Ollama installed locally is required to use the agentic features of Margarita (i.e. to run .mgx files). We're working on adding more llm backends currently.

Installation

Run the following command to install Margarita using uv:

uv tool install margarita

margarita use ollama

margarita run example.mgx

---

.mg — Prompt Templates

.mg files are Margarita templates. They render to plain Markdown and can be used anywhere Markdown is supported.

Hello World

// file: helloworld.mgx
@state name = "World"

<<
Hello, ${name}!
Welcome to Margarita templating.
>>

@effect run

Conditionals

if is_admin:
    << Welcome, Admin ${name}! >>
else:
    << Welcome, User ${name}! >>

Loops

<< # Items >>
for item in items:
    <<
    - ${item}
    >>

Range loops are also supported:

for i in range(3):
    << Step ${i} >>

Includes

Split prompts into reusable fragments and compose them:

// file: role.mg
<< You are a ${type} AI assistant. >>

// file: prompt.mg
[[ role type="helpful" ]]

if output_json:
    [[ json_output_format ]]

Metadata

---
title: "Greeting Template"
version: "1.0"
author: "Batman"
---

<<
Hello, ${name}!
Welcome to Margarita templating.
>>

CLI Reference

# Render a template
margarita render template.mg

# Render with inline JSON context
margarita render template.mg -c '{"name": "Alice"}'

# Render with a context file
margarita render template.mg -f context.json

# Render a directory of .mg files
margarita render templates/ -o output/

# Show metadata
margarita metadata template.mg

# Render and show metadata
margarita render template.mg --show-metadata

---

.mgx — Agent Templates

.mgx files extend .mg templates with agentic capabilities: Python imports, @state, @memory, @effect directives, and agent execution.

Run them with:

margarita run example.mgx

Note: Margarita's agent runner uses GitHub Copilot CLI. You will need it installed and configured.

Hello World Agent

---
description: Hello world agent template
---

<<
# Hello World

Tell the user Hello, and welcome them to Margarita!
>>

@effect run

The << >> block loads Markdown content into the agent's context. @effect run tells the agent to execute with the current context.

State

Define variables accessible during a run with @state:

@state count = 0

<< Set the count variable to 5. >>

@effect run

Memory

Persist variables across runs with @memory. Values are saved to memory.json at the end of each run and loaded at the start of the next:

@memory var favorite_color

<<
If favorite_color is not set, set it to "blue".
Otherwise log "The user's favorite color is ${favorite_color}".
>>

@effect run

Custom Tools

Register Python functions as LLM-callable tools with @effect tools:

from math import add, AddParams

<< Add 3 and 5. >>

@effect tools add(x: AddParams) => result

@effect run

Note: Tool params must be a valid Pydantic model.

Function Calls

Execute Python functions directly (without an LLM tool call) and save their result to state:

from my_module import compute

@effect func compute(x) => result

User Input

Prompt the user for input during a run:

@effect input "What is your favorite color?" => favorite_color

@effect log "The user's favorite color is ${favorite_color}."

Specify the Model

---
model: "gpt-4"
---

<< Your prompt here. >>

@effect run

---

Python Library

Install via pip, poetry, uv, or any package manager:

pip install margarita
poetry add margarita
uv add margarita

Basic Usage

from margarita.parser import Parser
from margarita.renderer import Renderer

template = """
<<
You are a helpful assistant.

Task: ${task}
>>
if context:
    <<
    Context:
        ${context}
    >>

<< Please provide a detailed response. >>
"""

parser = Parser()
metadata, nodes = parser.parse(template)

renderer = Renderer(
    context={"task": "Summarize the key points", "context": "User is researching AI agents"}
)

prompt = renderer.render(nodes)
print(prompt)

Composer

Use the Composer to build prompts from multiple template fragments:

from margarita.composer import Composer
from pathlib import Path

manager = Composer(Path("./templates"))

prompt = manager.compose_prompt(
    snippets=[
        "snippets/system_role.mg",
        "snippets/task_context.mg",
        "snippets/chain_of_thought.mg",
        "snippets/output_format.mg"
    ],
    context={
        "role": "data scientist",
        "user_name": "Bob",
        "task": "Analyze customer churn",
        "format": "JSON",
        "tone": "analytical"
    }
)

---

Documentation

Full documentation is available at https://banyango.github.io/margarita/latest

Development

This project uses uv for dependency management.

Setup

make install

Tests

make test

Code Quality

make format
make lint

Build

uv build

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Please make sure to:

• Update tests as appropriate
• Follow the existing code style
• Update documentation for any changed functionality

Authors

• Kyle Reczek - Initial work - Banyango

Changelog

See CHANGELOG.md for a history of changes to this project.

Support

If you encounter any problems or have questions, please open an issue.