GitAI 🤖

AI-powered commit message generator using local LLMs via Ollama. Generate meaningful, conventional commit messages automatically from your staged changes.

Features

• 🚀 Local AI: Uses Ollama for privacy and offline capability
• 📝 Conventional Commits: Follows industry-standard commit message format
• ⚡ Fast: Optimized for quick generation with model caching
• 🔧 Flexible: Works as standalone CLI or integrated git hooks
• 🎯 Smart: Analyzes actual code changes, not just file names
• ⚙️ Configurable: Customize models, prompts, and behavior

Installation

Prerequisites

• Ollama installed and running
• Git repository
• Go 1.21+ (for building from source)

Install GitAI

# Using go install
go install github.com/pesnik/gitai@latest

# Or download from releases
curl -L https://github.com/pesnik/gitai/releases/latest/download/gitai-linux-amd64 -o gitai
chmod +x gitai
sudo mv gitai /usr/local/bin/

Setup Ollama Model

# Download recommended model
ollama pull codellama:7b

# Or use a smaller model for faster generation
ollama pull llama3:8b

Quick Start

Standalone Usage

# Stage your changes
git add .

# Generate commit message
gitai generate

# Use the generated message
git commit -m "$(gitai generate)"

Git Hook Integration

# Install git hooks in current repository
gitai install

# Now git commit will automatically generate messages
git add .
git commit  # Opens editor with AI-generated message

Usage

Commands

gitai generate          # Generate commit message for staged changes
gitai install           # Install git hooks in current repo
gitai uninstall         # Remove git hooks
gitai config            # Show current configuration
gitai config set <key> <value>  # Set configuration option
gitai version           # Show version information

Configuration

# Set Ollama model
gitai config set model codellama:7b

# Set Ollama host (if not default)
gitai config set host http://localhost:11434

# Customize prompt template
gitai config set prompt "Custom prompt template..."

# Set max diff size (in lines)
gitai config set max-diff-lines 500

Configuration File

GitAI stores configuration in ~/.gitai/config.yaml:

model: "codellama:7b"
host: "http://localhost:11434"
timeout: 30
max_diff_lines: 500
prompt_template: |
  Analyze this git diff and generate a concise commit message following conventional commits format:
  
  {diff}
  
  Rules:
  - Start with type: feat, fix, docs, style, refactor, test, chore
  - Include scope in parentheses if applicable
  - Use present tense, imperative mood
  - Keep subject line under 72 characters
  - No period at the end of subject line
  
  Generate only the commit message, no explanation.

Examples

Generated Messages

# Adding a new feature
feat(auth): add OAuth2 login integration

# Fixing a bug
fix(parser): handle null values in JSON response

# Refactoring code
refactor(utils): extract string validation to separate function

# Updating documentation
docs(readme): add installation instructions for Windows

# Adding tests
test(auth): add unit tests for login validation

Workflow Integration

# Your typical workflow
git add src/auth.go
git commit
# Editor opens with: "feat(auth): add JWT token validation"
# Edit if needed, save and close

Advanced Usage

Custom Prompts

Create custom prompt templates for specific project needs:

gitai config set prompt "$(cat <<EOF
You are a senior developer writing commit messages for a React application.
Analyze this diff and create a commit message that:
- Follows conventional commits
- Mentions component names when relevant
- Includes performance impact if applicable

Diff:
{diff}

Generate only the commit message:
EOF
)"

Multiple Models

Switch between models for different scenarios:

# Fast model for quick commits
gitai config set model llama3:8b

# Detailed model for complex changes
gitai config set model codellama:13b

Integration with Git Aliases

Add to your .gitconfig:

[alias]
    ai = !gitai generate
    aic = !git add . && git commit -m "$(gitai generate)"

Troubleshooting

Common Issues

Ollama not responding

# Check if Ollama is running
ollama list

# Start Ollama if needed
ollama serve

Model not found

# Pull the model
ollama pull codellama:7b

# List available models
ollama list

Large diffs timing out

# Reduce max diff size
gitai config set max-diff-lines 200

# Or use a smaller/faster model
gitai config set model llama3:8b

Generated messages are poor quality

# Try a different model
gitai config set model codellama:13b

# Or customize the prompt
gitai config set prompt "Your custom prompt here..."

Development

Building from Source

git clone https://github.com/pesnik/gitai.git
cd gitai
go build -o gitai ./cmd/gitai

Running Tests

go test ./...

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'feat: add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

MIT License - see LICENSE file for details.

Acknowledgments

• Ollama for providing local LLM capabilities
• Conventional Commits for the commit message standard
• The Go community for excellent tooling and libraries

---

Made with ❤️ for developers who want better commit messages without sacrificing privacy