A collection of powerful CLI tools for automating development workflows and task management. These scripts are designed to be added to your system PATH for easy command-line access.

Repository: https://github.com/tom-doerr/path_scripts

Automates code improvements using AI agents in isolated git worktrees.

• Creates isolated git worktrees for safe experimentation
• Runs AI-powered code analysis and improvements
• Automatically merges changes back to main branch
• Handles merge conflicts intelligently
• Test-driven development workflow
• Exponential retry strategy for complex tasks

# Clone and install
git clone https://github.com/tom-doerr/path_scripts.git
cd path_scripts
pip install -r requirements.txt

# Add to PATH (optional)
ln -s $PWD/agent-aider-worktree ~/.local/bin/

# Basic usage
agent-aider-worktree "Add user authentication"

# With custom iterations and model
agent-aider-worktree --max-iterations 20 --model claude-3-opus "Refactor database code"

# From a subdirectory
agent-aider-worktree --exponential-retries "Fix login form validation"

Runs multiple AI agents in parallel for distributed code analysis.

• Parallel agent execution in tmux sessions
• Configurable models and iterations
• Session persistence and recovery
• Centralized management interface

# Start 3 parallel agents
aider_multi_agent -n 3

# Use specific model
aider_multi_agent -m claude-3-opus -n 2

# Kill all sessions
aider_multi_agent -k

Streamlined interface for Taskwarrior with smart filtering.

# Install to PATH
cp c ~/.local/bin/
chmod +x ~/.local/bin/c

# Configure task script path
export TASK_SCRIPT_PATH="/path/to/show_tw_tasks.py"

# Show pending tasks
c

# Filter by tag and report
c "+work" "active"

# Custom filters
c -n "+project:home" "next"

• Python 3.8+
• Git 2.25+
• tmux
• Taskwarrior (for task management)
• Required Python packages in requirements.txt

1. Fork the repository
2. Create a feature branch
3. Submit a pull request

MIT License - See LICENSE file for details