_     _                  _            _          
           | |   | |                | |          | |         
 _ __ _   _| |__ | |__   ___ _ __ __| |_   _  ___| | ___   _ 
| '__| | | | '_ \| '_ \ / _ \ '__/ _` | | | |/ __| |/ / | | |
| |  | |_| | |_) | |_) |  __/ | | (_| | |_| | (__|   <| |_| |   
|_|   \__,_|_.__/|_.__/ \___|_|  \__,_|\__,_|\___|_|\_\\__, |    
                                                        __/ |
                                                        |___/ 

                >(.)__ <(.)__ =(.)__
                 (___/  (___/  (___/ 

RubberDuck Shell is an interactive command-line tool that allows you to run Python scripts, diagnose errors using a variety of LLM models, and execute shell commands. It uses the traceback, and takes relevant code functions as context. Currently the default model is set to Llama-3-8B.

This has not been tested extensively. It was a quick fun project and I have not had the time to maintain it!

• An GROQ API key: follow the link to setup a GROQ api key

1. Clone the repository:

   git clone https://github.com/yourusername/rubberduck-shell.git
   cd rubberducky

2. Install required Python packages:

   pip install -r requirements.txt

1. Start the RubberDuck Shell:

   python rubberduck.py

2. Interact with the Shell:

   • Get help for commands:

     rubberduck>> help
     

   • Run a Python script:

     rubberduck>> python /path/to/your/script.py arg1 arg2 arg3
     

   • Diagnose the last error:

     rubberduck>> diagnose
     

   • Execute a shell command:

     rubberduck>> ls
     

   • Exit the shell:

     rubberduck>> exit
     

# test.py

import sys

def test(option):
    if option == "fail":
        rise = 1
        run = 2 * 0
        return rise/run
    else:
        return True

if __name__ == "__main__":
    option = sys.argv[1]
    hello = test(option)

• Currently the functionality for providing function code as context only works for .py files (ast module). There's a couple ideas here:
  • a more general purpose parse (ex. tree-sitter)
  • having intermediate steps and parse/extract function code using LLMs
• Code completion:
  • Take the solution given by the LLM and have the possibility to inject/replace the problematic code
• Beautify the layout:
  • creating something like a visual studio code extension? have some basic UI for it rather than a command line interface