GitWise Quick Reference

🚀 Setup

pip install pygitwise               # Basic installation
pip install "pygitwise[offline]"    # With offline model support
gitwise init                      # Configure AI backend

🤖 AI Backend Selection

# Environment variables
export GITWISE_LLM_BACKEND=ollama    # Default (recommended)
export GITWISE_LLM_BACKEND=offline   # Built-in model
export GITWISE_LLM_BACKEND=online    # OpenRouter API

# Quick switch
gitwise init  # Interactive reconfiguration

📝 Core Commands

Stage & Commit

gitwise add .                     # Interactive staging
gitwise add file1.py file2.py     # Stage specific files
gitwise commit                    # AI-generated commit message
gitwise commit --group            # Split into multiple commits

Context Management

gitwise set-context "description" # Set context for better AI suggestions
gitwise get-context               # View current branch context

Push & PR

gitwise push                      # Push with PR prompt
gitwise pr                        # Create PR with AI description
gitwise pr --labels               # Add suggested labels
gitwise pr --checklist            # Add context-aware checklist
gitwise pr --draft                # Create draft PR

Changelog

gitwise changelog                 # Generate changelog for release
gitwise changelog --auto-update   # Update unreleased section
gitwise setup-hooks               # Install pre-commit hook

Git Passthrough

gitwise status                    # git status
gitwise log --oneline -5          # git log
gitwise diff                      # git diff
gitwise <any-git-command>         # Direct passthrough

⚙️ Configuration

Ollama

export OLLAMA_MODEL=llama3        # Change model
export OLLAMA_URL=http://localhost:11434  # Custom server
ollama pull codellama             # Download new model

Offline

export GITWISE_OFFLINE_MODEL="TinyLlama/TinyLlama-1.1B-Chat-v1.0"

Online

export OPENROUTER_API_KEY="your_key"
export OPENROUTER_MODEL="anthropic/claude-3-haiku"

🔥 Pro Tips

  1. Group commits for refactoring: gitwise commit --group
  2. Quick PR with everything: gitwise pr --labels --checklist
  3. Check Ollama status: curl http://localhost:11434/api/tags
  4. Force offline mode: export GITWISE_LLM_BACKEND=offline
  5. Custom config path: export GITWISE_CONFIG_PATH=~/.myconfig.json

🆘 Troubleshooting

# Ollama not working?
ollama serve                      # Start Ollama service
ollama list                       # Check available models

# Permission issues?
chmod +x ~/.git/hooks/pre-commit  # Fix hook permissions

# Config issues?
rm ~/.gitwise/config.json         # Reset configuration
gitwise init                      # Reconfigure