Arduino Code Generator API

Overview

This project is a Flask-based API that generates fully functional and well-documented Arduino code based on user-provided project requirements. The system utilizes an AI model to analyze input parameters and generate code files, including a .ino sketch, a README.md documentation file, and optionally, a config.h configuration file.

Features

• Automatic Code Generation: Converts project specifications into complete Arduino code.
• Supports Multiple Components: Handles different sensors, actuators, and MCUs.
• Smart Defaults: Fills in missing values with the best options.
• Well-Documented Output: Generates comments and documentation for better understanding.
• Error Handling & Logging: Provides detailed error messages and logs responses for debugging.

Tech Stack

• Backend: Python (Flask)
• AI Model: OpenAI-compatible model via Groq API
• Environment Management: dotenv
• Logging: Python logging module

Setup Instructions

1. Clone the Repository

 git clone https://github.com/lily-osp/arduino-code-generator.git
 cd arduino-code-generator

2. Install Dependencies

Ensure you have Python 3 installed, then install the required packages:

 pip install -r requirements.txt

3. Configure Environment Variables

Create a .env file in the project root and set your Groq API key:

GROQ_API_KEY=your_api_key_here
FLASK_DEBUG=True  # Set to False in production

4. Run the Application

 python app.py

By default, the app runs on http://127.0.0.1:5000/.

API Endpoints

1. Generate Arduino Code

Endpoint: POST /generate

Request Body (JSON format):

{
  "Project name": "Smart Garden Monitor",
  "Definition/use case": "Monitor soil moisture and control irrigation",
  "Sensors": ["soil moisture", "DHT11"],
  "Actuators": ["water pump", "LCD"],
  "MCU": "Arduino Nano",
  "Other parameters": {
    "communication": "Serial",
    "control": "automatic",
    "timing": ["every 6 hours", "instant on low moisture"]
  }
}

Response (Example):

{
  "code.ino": "// Arduino code here...",
  "README.md": "# Project Documentation...",
  "config.h": "// Configuration constants..."
}

Logging & Debugging

• All logs are saved in app.log.
• If FLASK_DEBUG=True, API responses are also logged in debug_ai_response.json.

Deployment

For production, run the Flask app using Gunicorn:

 gunicorn -w 4 -b 0.0.0.0:8000 app:app

License

This project is licensed under the MIT License.