AI-Reqs: The Intelligent requirements.txt Generator

Tired of messy requirements.txt files? Frustrated that tools like pipreqs fail on your complex python projects and Jupyter Notebooks?

ai-reqs is a next-generation dependency scanner for Python that uses the power of Google's Gemini AI to solve the problems other tools can't. It intelligently scans your project, including .py files and .ipynb notebooks, and generates a clean, accurate requirements.txt file.

Why is ai-reqs better?

Installation

pip install ai-reqs

Quick Start

1. Get a Gemini API Key:

   • Go to Google AI Studio and create an API key.

2. Set the API Key (Optional): You can either set it as an environment variable (recommended) or pass it as an argument. On macOS / Linux:

   # Recommended: Set as an environment variable
   export GEMINI_API_KEY="YOUR_API_KEY_HERE"
   

   On Windows:

   # Recommended: Set as an environment variable
   set GEMINI_API_KEY="YOUR_API_KEY_HERE"
   

3. Run ai-reqs: Navigate to your project's root directory in your terminal and run the command:

   ai-reqs
   

   Alternatively, you can pass the key directly as an argument (avoiding step 2):

   ai-reqs --api-key YOUR_API_KEY
   

   That's it! A clean requirements.txt will be generated in your project directory.

Usage

The command-line interface is simple and straightforward.

Usage

usage: ai-reqs [-h] [--path PATH] [--api-key API_KEY]

Generate a requirements.txt file for a Python project using AI for dependency resolution.

options:
  -h, --help         show this help message and exit
  --path PATH        The path to the project directory to scan. Defaults to
                     the current directory.
  --api-key API_KEY  Your Gemini API key. Can also be set via the
                     GEMINI_API_KEY environment variable.

How It Works

The tool scans all Python scripts and Jupyter Notebooks in your project to find all import statements. It then uses a hybrid approach to find the correct package for each import:

1. It first checks your local Python environment's metadata.
2. For any import it can't resolve locally (like sklearn or cv2, which have different package names), it queries the Gemini API to get the correct pip package name.
3. Finally, it generates a requirements.txt file with the correct package names and their currently installed versions.

Contributing

Contributions are welcome! Please feel free to open an issue or submit a pull request.

License

This project is licensed under the MIT License.