How to set up Claude Code in iTerm2 to launch AI projects in one click

A ZDNet columnist has described his personal iTerm2 setup on macOS. With it, he can launch several projects running Claude Code, Anthropic's CLI tool for AI-assisted programming, in a single click, without typing commands by hand in each tab.

What's wrong with launching manually

Developers who work across several Claude Code repositories at once typically open a terminal, `cd` into the right folder, and start the agent manually in each project. With a dozen active projects, this becomes a chore that eats up time before the actual work even begins.

• Claude Code is Anthropic's CLI agent for writing and refactoring code directly in the terminal
• iTerm2 is a popular macOS terminal that supports profiles and hotkeys
• The setup relies on iTerm2 profiles that automatically open the right directory and start Claude Code

How iTerm2 and Claude Code work together

The solution uses iTerm2's built-in profiles. Each profile is tied to a specific project folder and launch command, so switching between tasks comes down to picking a profile or pressing an assigned key combination. The terminal opens a tab, moves into the repository directory, and starts Claude Code there. The agent is immediately ready to take prompts for that project.

This approach eliminates repetitive manual steps, namely typing `cd` and the launch command, and reduces friction when switching between multiple AI coding projects throughout the day.

What it means

As developers run CLI agents like Claude Code in parallel across multiple repositories, customizing the terminal environment is becoming as much a part of the everyday AI toolkit as choosing the model itself.