docify-ai 1.5.0

An AI-powered README.md generator for local projects.

Docify AI 🚀: Intelligent Documentation & Test Generation for Your Code

A versatile AI-powered command-line tool for instantly generating comprehensive README.md files, pytest tests, Dockerfiles, and GitHub Actions CI/CD workflows for your local projects, powered by Google Gemini and OpenAI GPT models.

Key Features ✨

• 🤖 AI-Powered Content Generation: Leverages large language models (Google Gemini and OpenAI GPT) to write human-quality README.md files, pytest tests, Dockerfiles, and GitHub Actions CI/CD workflows.
• 💡 Versatile Output Formats: Generate comprehensive project documentation, runnable unit tests, containerization configurations, or automation workflows.
• 📂 Intelligent Project Analysis: Scans your project directory, intelligently ignoring irrelevant files and directories (configurable with --ignore-dirs and --ignore-exts), to deeply understand its purpose, technologies, and structure.
• ⚙️ Flexible AI Client Selection: Seamlessly choose between Google Gemini and OpenAI GPT models for content generation directly from the command line.
• 🔑 Robust API Key Management: Supports API keys via environment variables or direct command-line arguments for both Gemini and OpenAI for enhanced security and convenience.
• 🚀 Fast Project Scaffolding: Bootstrap new Python projects with an intelligent, AI-generated initial structure using the --init option.
• 📝 Docstring Automation: Automatically generate and insert Google-style docstrings for Python modules, classes, and functions using the --docstring option.
• ⏱️ Rapid Generation: Transform a nascent project or an existing codebase into a fully documented, tested, and deployable application in moments.
• 🔧 Highly Customizable: Specify project paths, output file names, and tailor content generation with various command-line options.

Installation 📦

Docify-AI is available on PyPI and can be installed on any system with Python 3.8 or newer using pip.

pip install docify-ai

Usage 🚀

Using Docify-AI is straightforward, involving a one-time API key setup and a simple command execution.

1. Set Your API Key

Docify-AI requires an API key for the chosen AI model (Google Gemini or OpenAI). You can obtain a free key from Google AI Studio for Gemini, or from the OpenAI platform for OpenAI.

Once you have your key, set it as an environment variable:

• For Google Gemini: GEMINI_API_KEY
• For OpenAI: OPENAI_API_KEY

For macOS / Linux (bash/zsh):

export GEMINI_API_KEY='your-gemini-secret-api-key'
# OR
export OPENAI_API_KEY='your-openai-secret-api-key'

(To make this permanent across terminal sessions, add the line to your shell's configuration file, e.g., ~/.zshrc or ~/.bashrc.)

For Windows (PowerShell):

$Env:GEMINI_API_KEY="your-gemini-secret-api-key"
# OR
$Env:OPENAI_API_KEY="your-openai-secret-api-key"

Alternatively, you can pass the API key directly via the --key argument, which will override the environment variable.

2. Run Docify-AI

Navigate to the root directory of the project you wish to document and simply run the docify command. By default, it uses the Gemini model to generate a README.

docify

The tool will scan the current directory and generate a README.md file with AI-powered content.

Command-Line Options

You can customize the behavior of Docify-AI using the following command-line arguments:

• --path <directory> / -p <directory>: Specifies the root directory of the project to be analyzed. Defaults to the current working directory (.).
• --output <filename_or_dir> / -o <filename_or_dir>: Defines the name of the output file or directory. Defaults: README.md for docs, tests/ for tests, Dockerfile for Docker, .github/workflows/ci.yml for GitHub Actions, or the modified file for docstring generation.
• --client <openai|gemini> / -c <openai|gemini>: Choose the AI client to use for generation. Options are openai or gemini (default: gemini). Case-insensitive.
• --key <your-api-key> / -k <your-api-key>: Provide your API key directly. This will take precedence over environment variables.
• --test / -t: Generate pytest test files instead of a README.
• --docker: Generate a Dockerfile for the project.
• --gha: Generate a GitHub Actions (CI/CD) workflow.
• --init <project_name>: Bootstrap a new Python project with the given project name, creating a basic scaffold.
• --docstring <file_path>: Add Google-style docstrings to a specified Python file (relative path from current directory).
• --ignore-dirs <dir1> <dir2>...: A space-separated list of directory names to ignore during scanning (e.g., .git __pycache__ node_modules).
• --ignore-exts <ext1> <ext2>...: A space-separated list of file extensions to ignore (e.g., .tmp .pyc .log).

Examples:

1. To document a project located at /path/to/my-other-project and save the output to PROJECT_DOCS.md using the default Gemini model:

   docify --path /path/to/my-other-project --output PROJECT_DOCS.md
   

2. To use the OpenAI client for README generation:

   docify --client openai
   

3. To use the OpenAI client and provide the API key directly (overriding any environment variable):

   docify --client openai --key sk-YOUR_OPENAI_API_KEY
   

4. To generate pytest tests for the current project using the Gemini client, ignoring dist and build directories:

   docify --test --client gemini --ignore-dirs dist build
   

5. To generate a Dockerfile for the current project:

   docify --docker
   

6. To generate a GitHub Actions workflow for the current project:

   docify --gha
   

7. To initialize a new Python project named my_new_app:

   docify --init my_new_app
   

8. To add docstrings to my_module.py in the current directory:

   docify --docstring my_module.py
   

1.5.0 - 2025-09-05

• 📝 Added --docstring option to automatically generate PEP 257–compliant docstrings for Python functions, classes, and modules.

1.4.0 - 2025-08-28

• Added --docker , --gha and --init options, to generate the project aware Dockerfile, Github Actions worflow Yaml file and to generate a project structure for a python project.

1.3.0 - 2025-08-22

• Added --ignore-dirs and --ignore-exts option, to ignore the irrelevant or unwanted directories and files to generate the docs/readme or tests.

1.2.0 - 2025-08-21

• Added -t or --test option, to generate pytest supported test directory including test modules for all the features of your project.

1.1.1 - 2025-08-20

• fixed project description not visible issue

1.1.0 - 2025-08-20

• ✨ Added support for OpenAI + Gemini clients
• 🛠 Added --key argument to provide api key, without setting it as environment variable.
• 🔍 Case-insensitive --client option

1.0.0 - 2025-08-15

• 🚀 Initial release