llm-autodocs 🪄

Overview 🌞

autodocs is a Python-based command-line tool for automatically generating documentation for git-tracked Python files in a project. It leverages LLMs to generate documentation automatically and asynchronously.

Features ⚡️

• Git Integration: Only selects Python files that are tracked by Git within a specified directory.
• Custom File Selection: Offers the flexibility to specify files to include or exclude in the documentation process.
• LLM Support: Currently supports OpenAI gpt models
• Asynchronous Execution: Improves performance by handling multiple files concurrently.

Quickstart 🚴‍♂️

Install the library:

pip install llm-autodocs

Navigave to your git project and run:

cd my_project/
autodocs --directory .

This will run in the root of your project and include all tracked .py files. By default it uses gpt-3.5-turbo.

You will be prompted to review the files affected before continuing.

Examples

Specify a specific directory:

autodocs --directory ./src

Specify a different openai model (can be any of the gpt-3.5* or gpt-4* patterns):

autodocs --documenter gpt-4

Allow only certain file patterns:

autodocs --allowed-files utils.py data_model.py

Exclude certain file patterns:

autodocs --ignored-files __init__.py