What are Git Worktrees?
Git worktrees allow you to have multiple working directories from the same repository, enabling parallel development without the complexity of multiple clones or constant branch switching.
Traditional vs. Worktree Workflow
Traditional: Switch branches → Handle PR feedback → Switch back → Lose context
Worktree: Main directory (new features) + Review directory (PR feedback) running simultaneously
Benefits for AI-Assisted Development
- Parallel PR Reviews: Handle CodeRabbit feedback without interrupting main development
- Hotfix Support: Quickly fix critical issues while continuing feature work
- Experimental Features: Test new approaches without affecting stable development
- Context Preservation: No need to stash/commit incomplete work when switching focus
- Phase Independence: Work on multiple development phases simultaneously
💡 AI Integration: Your AI assistant will analyze development phases and suggest when parallel workflows would be beneficial, always asking for your permission before making changes.
Setting Up Git Worktrees
1
One-Time Setup
Create dedicated worktrees for different types of work:
# Create worktree for PR reviews and hotfixes
git worktree add ../myproject-reviews main
# Create worktree for experimental features
git worktree add ../myproject-experiments main
2
Verify Setup
# List all worktrees
git worktree list
# Should show:
# /path/to/myproject abc123 [main]
# /path/to/myproject-reviews def456 [main]
3
Directory Structure
📁 Projects/
├── 📁 myproject/ # Main development
│ ├── 📄 Plan.md
│ └── 📁 src/
└── 📁 myproject-reviews/ # PR feedback & hotfixes
├── 📄 Plan.md
└── 📁 src/
⚠️ Important: Each worktree is a complete working directory. Changes in one worktree don't affect others until you commit and merge.
Parallel Development Workflows
1. PR Review Workflow
# Main directory: Continue Phase 3 development
cd /path/to/myproject
git checkout phase-3-feature
# Review directory: Handle CodeRabbit feedback for Phase 2
cd /path/to/myproject-reviews
git checkout phase-2-feature
# Fix review comments, commit, push
# Monitor PR status from either directory
gh pr view --comments
gh pr status
2. Hotfix Workflow
# Critical bug reported in production
cd /path/to/myproject-reviews
git checkout main
git checkout -b hotfix-critical-bug
# Fix issue, test, commit
git push -u origin hotfix-critical-bug
gh pr create --title "Hotfix: Critical bug"
# Main development continues uninterrupted
cd /path/to/myproject
# Continue feature work...
3. Phase Independence Analysis
Your AI assistant will analyze development phases using these criteria:
- Does this phase enhance/extend previous phase functionality?
- Do they share UI components or settings interfaces?
- Are there logical integration points for better UX?
- Do they use related frameworks or services?
💡 AI Decision Making: Based on this analysis, your AI assistant will suggest whether phases can be developed in parallel and ask for permission to update your development plan.
4. Cleanup Workflow
# After PR is merged, clean up
cd /path/to/myproject-reviews
git checkout main
git pull origin main
git branch -d phase-2-feature
# Remove worktree when no longer needed
git worktree remove /path/to/myproject-reviews
Real-World Examples
Example 1: Mobile App Development
# Main: Working on Phase 4 (Deep Session Timer)
cd /Users/dev/DeepRoot
git checkout phase-4-deep-session-timer
# Reviews: Handling CodeRabbit feedback for Phase 3 (Movement Feature)
cd /Users/dev/DeepRoot-reviews
git checkout phase-3-movement-feature
# Apply feedback, commit, push
Example 2: Web Application
# Main: Building new authentication system
cd /Users/dev/webapp
git checkout phase-2-auth-system
# Experiments: Testing new UI framework
cd /Users/dev/webapp-experiments
git checkout experiment-new-ui
# Prototype new approach without affecting main work
Example 3: API Development
# Main: Implementing user management endpoints
cd /Users/dev/api-project
git checkout phase-3-user-management
# Reviews: Fixing security review feedback
cd /Users/dev/api-project-reviews
git checkout phase-2-data-validation
# Address security concerns, update tests
Common Parallel Scenarios
- Independent UI Components: Frontend components that don't share state
- Separate Microservices: Different services with minimal dependencies
- Documentation vs. Code: Writing docs while implementing features
- Testing Strategies: Developing tests for previous phases while building new features
- Performance Optimization: Optimizing existing code while adding new functionality
⚠️ When NOT to Use Parallel Development:
- Phases with heavy dependencies
- Shared database schema changes
- Core architecture modifications
- Breaking API changes
Integration with AI Development Plans
Your AI assistant will automatically detect opportunities for parallel workflows and:
- Analyze phase dependencies in your DEVELOPMENT_PLAN.md
- Suggest specific parallel workflow opportunities
- Ask for permission before updating your development plan
- Provide specific setup instructions when approved