Docify-AI

An intelligent command-line interface (CLI) that leverages AI to automate documentation, testing, and project scaffolding for local software projects.

Key Features

• AI-Powered Documentation Generation: Automatically create detailed and domain-specific README.md files for any project.
• Automated Test Suite Generation: Generate runnable pytest test files covering various scenarios.
• Dockerfile Generation: Create optimized Dockerfiles for containerizing your applications.
• GitHub Actions Workflow Generation: Develop YAML configurations for CI/CD pipelines (e.g., build, test, deploy).
• Project Scaffolding: Bootstrap new Python projects with a predefined structure.
• Docstring Generation: Automatically add PEP 257–compliant docstrings to Python functions, classes, and modules.
• Jupyter Notebook Generation: Kickstart data analysis or ML projects with a basic Jupyter Notebook including data loading, EDA, and model training/evaluation pipelines.
• Model Card Generation: Create MODEL_CARD.md for ML/AI projects, detailing model information, intended use, datasets, and ethical considerations.
• Flexible AI Clients: Support for both Google Gemini and OpenAI GPT models.
• Intelligent File Scanning: Ignores specified directories and file extensions to focus on relevant code.
• Dataset Schema Extraction: Automatically infers schema and samples from common data formats (.csv, .json, etc.) to inform AI generations for data-centric tasks.

Installation

Docify-AI can be installed directly from PyPI using pip:

pip install docify-ai

Usage

The docify command is your entry point to all functionalities.

API Key Setup

You must set your GEMINI_API_KEY or OPENAI_API_KEY environment variable or provide it via the --key argument.

# Example for setting environment variable (Linux/macOS)
export GEMINI_API_KEY='YOUR_GEMINI_API_KEY'
# or for OpenAI
export OPENAI_API_KEY='YOUR_OPENAI_API_KEY'

Global Options

• --path, -p: Root directory of the project (default: current directory . ).
• --output, -o: Custom output file/folder name.
• --client, -c: Choose AI client (openai or gemini, default: gemini). Case-insensitive.
• --key, -k: Provide API key directly (overrides environment variable).
• --ignore-dirs: Space-separated list of directories to ignore during scanning.
• --ignore-exts: Space-separated list of file extensions to ignore during scanning.

CLI Commands

1. Generate README.md (Default Action)

If no specific action flag is provided, docify generates a README.md for the project.

docify -p /path/to/your/project
# Output will be saved to /path/to/your/project/README.md

2. Generate Pytest Test Files

Creates a tests/ directory with pytest compatible test modules.

docify --test -p /path/to/your/project
# Output: tests/ directory within your project, e.g., tests/test_main.py

3. Generate Dockerfile

Creates a Dockerfile for your project.

docify --docker -p /path/to/your/project
# Output: Dockerfile within your project

4. Generate GitHub Actions Workflow

Creates a .github/workflows/ci.yml file for CI/CD.

docify --gha -p /path/to/your/project
# Output: .github/workflows/ci.yml within your project

5. Bootstrap a New Python Project

Generates a basic Python project structure.

docify --init my_new_project
# Output: A new directory named 'my_new_project' with basic Python project files.

6. Add Docstrings to a Python File

Inserts Google-style docstrings into a specified Python file.

docify --docstring /path/to/your/project/my_module.py
# The specified Python file will be updated in place.

7. Generate a Jupyter Notebook

Creates a starter Jupyter Notebook, especially useful for data-centric projects.

docify --notebook -p /path/to/your/data_project
# Output: notebook.ipynb within your project.

8. Generate a Model Card

Creates a MODEL_CARD.md for AI/ML projects.

docify --model-card -p /path/to/your/ml_project
# Output: MODEL_CARD.md within your project.