ATX - Aiello ToolX

Build standalone, LLM-ready tools from Python functions. Package your tools as portable executables that can be distributed and deployed anywhere.

What is ATX?

ATX (Aiello ToolX) is a framework for creating function tools that Large Language Models can discover and execute. It transforms Python functions into:

• Standalone executables - Single binary files with no external dependencies
• LLM-compatible - Self-describing tools with JSON schema definitions
• Distributable - Share tools across teams and environments
• Version-controlled - Manage multiple versions of the same tool

Quick Start

Installation

pip install atx-py

Create Your First Tool

Create a file my_tool.py:

from atx import AielloToolx
from typing import Any, Optional

class GreetTool(AielloToolx):
    def name(self) -> str:
        return "greet_user"

    def description(self, context: Optional[str] = None) -> str:
        return "Greets a user by name"

    def parameters(self, context: Optional[str] = None) -> dict[str, Any]:
        return {
            "type": "object",
            "properties": {
                "name": {
                    "type": "string",
                    "description": "Name of the person to greet"
                }
            },
            "required": ["name"]
        }

    def run(
        self, arguments: Optional[str] = None, context: Optional[str] = None
    ) -> str:
        import json
        args = json.loads(arguments) if arguments else {}
        name = args.get("name", "stranger")
        return f"Hello, {name}!"

Build the Tool

atx build my_tool.py \
  --tool-name greet_user \
  --tool-version 1 \
  --output-dir ./dist

Run the Tool

./dist/greet_user/1/tool --arguments '{"name": "Alice"}'

Output:

Hello, Alice!

Features

Type-Safe Parameters

Define parameters using JSON Schema with full type safety:

def parameters(self, context: Optional[str] = None) -> dict[str, Any]:
    return {
        "type": "object",
        "properties": {
            "location": {"type": "string"},
            "days": {"type": "integer", "enum": [1, 5, 10]}
        },
        "required": ["location"]
    }

Environment Variable Support

Securely handle API keys and sensitive data:

import os

def run(self, arguments: Optional[str] = None, context: Optional[str] = None) -> str:
    api_key = os.getenv("MY_API_KEY")
    if not api_key:
        raise ValueError("MY_API_KEY environment variable is required")
    # Use api_key...

Versioned Deployment

Deploy multiple versions side-by-side:

dist/
├── my_tool/
│   ├── 1/
│   │   └── tool
│   └── 2/
│       └── tool

Documentation

For comprehensive documentation, see:

• Documentation Home - Complete documentation index
• Getting Started Guide - Detailed tutorial
• Core Concepts - Understanding the framework
• API Reference - Complete API documentation

Examples

The project includes several prebuilt examples:

• hello_world - Minimal example
• get_maps_geocode - Google Maps geocoding
• get_weather_forecast_daily - Weather forecasts

Build and run examples:

# Build hello world
atx build atx/prebuilds/hello_world.py --tool-name hello_world --tool-version 1

# Run it
./dist/hello_world/1/tool

Development

Prerequisites

• Python 3.11+
• Poetry (for dependency management)

Setup

# Clone the repository
git clone <repository-url>
cd atx

# Install dependencies
poetry install

# Set up environment variables
cp .env.example .env

Running Tests

poetry run pytest

License

Private - Aiello Inc.

Contributing

This is a private project. For questions or contributions, please contact the Aiello development team.