Hello blog! I am excited to introduce my latest project, a CLI tool called addcom.

What is addcom?

addcom is a command-line tool created to help coders annotate their source code. To use it, you need to provide a relative or absolute path to your file, and addcom will analyze the code and generate comments using a Large Language Model's chat completion.

See a demo of the functionality on YouTube: addcom-demo

How I built it

I made addcom as part of my Open Source course assignment, where we were tasked to create a CLI tool that developers to leverage any OpenAI-compatible Chat Completion API endpoint to transform files in a helpful way.

I had never created a command-line interface tool before, so naturally, figuring out the right approach took some exploration.

From the start, I knew that I wanted to write addcom in Python. After watching a few different "build a CLI tool with Python" YouTube tutorials, taking a peek into my classmates' repositories and reading through the Release 0.1 wiki page, I finally settled on the right toolset:

• Typer - a framework created for making CLI applications.
• OpenAI chat completion API with Groq API endpoint - for generating comments using a Large Language Model.
• Setuptools - to package and distribute the tool.

Usage and Functionality

Setup Instructions

Make sure that you have Python installed on your system (you can download it here: https://www.python.org/downloads/).

1. Clone my GitHub repo. After cloning cd into the project folder and run:

pip install .

2. Create an account and generate the API key here: https://console.groq.com/

By default, addcom uses the Groq API endpoint for chat completion. However, you can specify a custom endpoint using the --base-url or -u flag option. (If you do this, make sure to obtain an appropriate API key and specify the model supported by the chosen provider using the --model/ -m option).

3. Set the API key

To do this you can expose the API key to the terminal environment:

Command Prompt

   set GROQ_API_KEY=your_api_key_here

Powershell

  $env:ADDCOM_API_KEY="your_api_key_here"

Bash

  $env:ADDCOM_API_KEY="your_api_key_here"

or provide the key using the --api-key/ a option flag.

4. Congrats! addcomis ready to use. To add comments to your source code files, simply run the command along with the file paths.

 addcom [OPTIONS] FILE_PATH(S)...

Features

Apart from its core functionality (utilizing Groq API and Llama3 to generate comments and printing the annotated code to standard output) addcom supports a number of additional features:

• Displaying a comprehensive help message with the --help flag

 addcom --help

• Showing the current tool version using --version / -v

 addcom --version

• Saving the commented code to the specified file using --output/ -o

If multiple files are specified, the commented source code from all files will be combined and saved into a single output file.

 addcom examples/test.py -o commented_test.py

• Providing the API key directly via cli option --api-key/ -a.

Passing the API key using this option will override the API key that was exposed to the terminal.

 addcom --api-key "your_api_key" examples/test.py

• Specifying an API endpoint (--base-url/ -u) and Large Language Model (--model/ -m) to be used for generating the comments

If you decide to use a custom API endpoint, make sure to obtain an appropriate API key and specify a Large Language Model supported by the API of your choice.

You can find models compatible with the default API endpoint here: https://console.groq.com/docs/models

Using OpenRouter API as base URL and Meta-Llama-3.1-8B as a model of choice:

  addcom -u https://openrouter.ai/api/v1 -a "your_api_key" -m meta-llama/llama-3.1-8b-instruct:free examples/test.py

Thank you for sticking this far!

Check out addcom on GitHub and feel free to contribute or give feedback!

pip install .