CLI implementation of Codex for explaining shell commands

Note: This script requires a valid OpenAI API key with access to the Codex (code-davinci-002) model.

1. Clone the repo
2. Install dependencies
   • pip install -r requirements.txt
3. Build the package
   • python -m build
4. Install the package
   • pip install dist/codex.sh-0.1-py3-none-any.whl
5. Set up your API key
   • echo "OPENAI_API_KEY=your_key_here" > .env
6. Run the package
   • codex

1. Set up your API key
   • echo "OPENAI_API_KEY=your_key_here" > .env
2. Install the package
   • pip install codex.sh
3. Run the package
   • codex

usage: codex [-h] [-t TEMP] [-n NUM] [-w] query [query ...]

OpenAI Codex CLI cheatsheet

positional arguments:
  query                 query to search for

optional arguments:
  -h, --help            show this help message and exit
  -t TEMP, --temp TEMP  Codex model temperature (randomness)
  -n NUM, --num NUM     Number of codex predictions to return
  -w, --wordcount       Print word count of prompt file for debugging
                        purposes

This project is licensed under the MIT License - see the LICENSE file for details.

• Thanks to OpenAI for making this possible!
• Thanks to Rich for CLI formatting!