Git-AutoCommit 🤖

AI-powered git commit message generator using Google's Gemini AI. Automatically create professional, conventional commit messages from your staged changes.

✨ Features

• 🤖 AI-Powered: Uses Google Gemini to analyze your changes and generate meaningful commit messages
• 📝 Conventional Commits: Follows the Conventional Commits standard
• ⚡ Easy to Use: Simple CLI with multiple modes (auto, interactive, dry-run)
• 🎯 Smart Analysis: Intelligently summarizes large diffs to focus on key changes
• 🔧 Configurable: Customize behavior with YAML config files
• 🚀 Quick Setup: One-command installation with automated scripts

🚀 Quick Start

Prerequisites

• Python 3.7 or higher
• Git
• Google Gemini API key

Installation

Option 1: Automated Installation (Recommended)

Cross-Platform (Python):

git clone https://github.com/LemonMantis5571/Git-AutoCommit.git
cd Git-AutoCommit
python install.py

Linux/Mac (Shell Script):

git clone https://github.com/LemonMantis5571/Git-AutoCommit.git
cd Git-AutoCommit
chmod +x install.sh
./install.sh

The installation script will:

• Check for Python and Git
• Install the package and dependencies
• Set up your API key (environment variable)
• Let you choose a Gemini model (gemini-2.5-flash, gemini-2.5-pro, etc.)
• Create the git aic alias (optional)
• Show usage instructions

After installation, use: python -m git_suggest

Option 2: Manual Installation

# Clone the repository
git clone https://github.com/LemonMantis5571/Git-AutoCommit.git
cd Git-AutoCommit

# Install the package
pip install -e .

# Set your API key (choose one method)
# Windows:
setx api_key "YOUR_GEMINI_API_KEY"

# Linux/Mac (add to ~/.bashrc or ~/.zshrc):
export api_key="YOUR_GEMINI_API_KEY"

# Optional: Create git alias
git config --global alias.aic '!python -m git_suggest'

Option 3: Install from GitHub (Coming Soon)

pip install git+https://github.com/LemonMantis5571/Git-AutoCommit.git

📖 Usage

Basic Usage

# Stage your changes
git add .

# Generate and commit with AI message
python -m git_suggest

CLI Options

python -m git_suggest --help              # Show help message
python -m git_suggest --version           # Show version
python -m git_suggest --dry-run           # Preview message without committing
python -m git_suggest --interactive       # Review/edit message before committing
python -m git_suggest --config PATH       # Use custom config file
python -m git_suggest --verbose           # Enable verbose output

Using the Git Alias

If you set up the git aic alias during installation:

git add .
git aic                         # Same as python -m git_suggest

Interactive Mode

Interactive mode lets you review, edit, or regenerate the message before committing:

python -m git_suggest --interactive

You'll see:

============================================================
Generated commit message:
============================================================

  feat: add user authentication endpoint

============================================================

Options: [c]ommit, [e]dit, [r]egenerate, [a]bort:

• c: Commit with the generated message
• e: Edit the message manually
• r: Ask AI to generate a new message
• a: Abort without committing

Dry Run Mode

Preview the AI-generated message without committing:

python -m git_suggest --dry-run

Output:

feat: add user authentication endpoint

⚙️ Configuration

Create a .gitcommit.yml file in your home directory (~/.gitcommit.yml) or project root:

# Google Gemini model to use
model: gemini-2.0-flash-exp

# Maximum diff lines before summarization
max_diff_lines: 300

# Environment variable name for API key
api_key_env: api_key

See .gitcommit.yml.example for more details.

📝 Commit Message Format

Generated messages follow the Conventional Commits specification:

Type	Description	Example
feat:	New features	feat: add user login endpoint
fix:	Bug fixes	fix: resolve null pointer in auth
docs:	Documentation	docs: update API documentation
style:	Code style/formatting	style: format code with prettier
refactor:	Code refactoring	refactor: simplify user service
test:	Tests	test: add unit tests for auth
chore:	Maintenance	chore: update dependencies

🔧 Troubleshooting

"api_key environment variable not set"

Solution:

1. Make sure you've set the environment variable
2. Windows: Restart your terminal after running setx
3. Linux/Mac: Run source ~/.bashrc or restart terminal

Verify it's set:

# Windows (PowerShell)
echo $env:api_key

# Linux/Mac
echo $api_key

"Error calling Gemini API"

Possible causes:

• Invalid API key
• No internet connection
• API quota exceeded

Solution: Verify your API key at Google AI Studio

"No staged changes found"

Solution: Stage files before running:

git add <files>
git-suggest

Command not found: git-suggest

Solution: Make sure the package is installed:

pip install -e .

Or use the full Python module path:

python -m git_suggest

🎯 Examples

Example 1: Quick Commit

$ git add src/auth.py
$ python -m git_suggest
[main abc1234] feat: add JWT authentication
 1 file changed, 45 insertions(+)

Example 2: Review Before Committing

$ git add .
$ python -m git_suggest --interactive
============================================================
Generated commit message:
============================================================

  refactor: restructure database models

============================================================

Options: [c]ommit, [e]dit, [r]egenerate, [a]bort: c
[main def5678] refactor: restructure database models
 3 files changed, 120 insertions(+), 45 deletions(-)

Example 3: Preview Only

$ git add README.md
$ python -m git_suggest --dry-run
docs: update installation instructions

🤝 Contributing

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

📄 License

This project is open source and available under the MIT License.

🙏 Acknowledgments

• Powered by Google Gemini AI
• Follows Conventional Commits specification

🔗 Links

• GitHub Repository
• Issue Tracker
• Get Gemini API Key

---

Made by LemonMantis5571