About Conductor
Conductor runs multiple Claude Code agents in parallel with isolated workspaces on your Mac. It’s built for developers and teams using Claude Code to orchestrate AI agents and track who’s working, who’s stuck, and what’s changing.
Key Features
In practice, Conductor helps you run parallel agents locally with clear workspace isolation and automatic git worktree handling, so you can experiment at scale without corrupting your repo:
Isolated Claude Code Workspaces
Each Claude Code you spin up runs in its own isolated workspace on your Mac.
Local repo cloning and macOS execution
Conductor clones your repository and runs agents entirely on your Mac.
Git worktree management
Handles all git worktree setup and management behind the scenes.
Agent orchestration and monitoring
Coordinates multiple Claude Code agents and provides at-a-glance visibility into who’s working, who’s stuck, and what’s changed.
Summary
Best for developers and teams testing Claude Code agents, AI researchers prototyping agent orchestration, and operations teams running reproducible local experiments.
