Auto Commit Bot 🤖

A Python package that automates Git commit message generation using Large Language Models (LLM). It analyzes your git diff and generates structured, meaningful commit messages following the Conventional Commits format.

Features ✨

• 🔄 Automatic commit message generation
• 🎯 Support for multiple LLM providers (OpenAI, Hugging Face, Local models)
• 📝 Conventional Commits format
• 🔍 Dry-run mode for message preview
• ⚙️ Configurable settings

Installation 📦

1. Install PyTorch

First, install PyTorch according to your CUDA version:

1. Visit https://pytorch.org
2. Select your preferences:
   • PyTorch Build: Stable
   • Your OS
   • Package Manager: Pip
   • CUDA Version: (Check your CUDA version or select CPU only)
3. Run the provided command, for example:

# For Windows with CUDA 11.8
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

# For CPU only
pip3 install torch torchvision torchaudio

2. Install Auto Commit Bot

# Clone the repository
git clone https://github.com/JuiHsuanLee/auto-commit-bot.git
cd auto-commit-bot

# Install dependencies and package
pip install -e .

3. Verify Installation

# Run using Python module
python -m auto_commit_bot --help

# Or use the command directly (if Python Scripts directory is in PATH)
auto-commit --help

Quick Start 🚀

1. Configure your preferred LLM provider:

# For OpenAI
auto-commit configure --provider openai --api-key your-api-key

# For Hugging Face
auto-commit configure --provider huggingface --api-key your-api-key --model your-model-name

# For local models
auto-commit configure --provider local --model path/to/your/model

2. Stage your changes:

git add .  # or stage specific files

3. Generate commit message and commit:

auto-commit commit

Usage 📖

Basic Commands

# Generate commit message and commit staged changes
auto-commit commit

# Preview commit message without committing (dry-run)
auto-commit commit --dry-run

# Stage all changes and commit
auto-commit commit --stage-all

# Override LLM provider for a single commit
auto-commit commit --provider openai

Configuration

The tool can be configured using the configure command or by creating a .acbconfig file:

# Set LLM provider
auto-commit configure --provider openai

# Set API key
auto-commit configure --api-key your-api-key

# Set model
auto-commit configure --model gpt-3.5-turbo

Example .acbconfig:

llm_provider: openai
openai_api_key: your-api-key
openai_model: gpt-3.5-turbo
commit_format: conventional

Custom Prompt Templates 📝

You can customize the prompt template by creating a file and setting its path in the configuration:

auto-commit configure --prompt-template path/to/template.txt

Development 🛠

1. Clone the repository
2. Install dependencies:

poetry install

3. Run tests:

poetry run pytest

Contributing 🤝

Contributions are welcome! Please feel free to submit a Pull Request.

License 📄

MIT License

Examples 📝

1. Basic Usage

from auto_commit_bot.llm_utils import LLMProvider
from auto_commit_bot.git_utils import get_git_diff

# Initialize the LLM provider
llm = LLMProvider()

# Get git diff
diff = get_git_diff(staged=True)

# Generate commit message
message = llm.generate_commit_message(diff)
print(f"Generated commit message: {message}")

2. Custom Configuration

from auto_commit_bot.config import Config

# Initialize configuration
config = Config()

# Set up Hugging Face API
config.set("provider_type", "api")
config.set("model_name", "gpt2")
config.set("huggingface_api_key", "your-api-key")
config.save()

# Use local model
config.set("provider_type", "local")
config.set("model_name", "path/to/your/model")
config.save()

3. Custom Prompt Template

# Create a custom prompt template
template = """
Analyze the following git diff and generate a commit message.
Focus on the main changes and keep it concise.

Git diff:
{diff}

Generate a commit message following this format:
type(scope): description
"""

# Save the template
with open("custom_template.txt", "w") as f:
    f.write(template)

# Configure to use the custom template
config.set("prompt_template", "custom_template.txt")
config.save()

4. Command Line Usage

# Stage and commit with default settings
git add .
auto-commit commit

# Preview commit message
auto-commit commit --dry-run

# Use specific provider
auto-commit commit --provider api

# Configure settings
auto-commit configure --provider api --model gpt2 --api-key your-api-key