AI Briefing
KOSign in

Farol Launches: Open-Source macOS Terminal for Managing Parallel Claude Code and Codex Agents

·2026.10.10 19:38

Key point

Farol is a free, MIT-licensed macOS terminal built on libghostty that tracks agent status via a sidebar and menu bar.

1 / 7

Details

Farol, a new macOS terminal application, launched today to help developers manage multiple AI coding agents simultaneously. Designed specifically for running Claude Code and Codex in parallel, it addresses the workflow pain point of constantly switching tabs to check which agent is stuck on a permission prompt or waiting for input.

Core Functionality

The terminal provides a visual status system where each session is marked as working, waiting for you, or done. This status is visible in a sidebar within the app and mirrored in the macOS menu bar or notch, allowing users to monitor agents while working in other applications. Farol does not wrap the agents; instead, it relies on the agents' native hooks to report their state.

Workflow Integration

Farol integrates directly with development workflows:

  • Task Creation: Users can start a new task by describing it, selecting a GitHub issue, or picking a Jira ticket. The app automatically creates a new git worktree on a fresh branch and starts the agent with the task as the initial prompt.
  • Review and Commit: Changes from a session are displayed as a single diff, enabling users to commit, push, and open a pull request directly within the terminal window.
  • Conflict Resolution: A three-column merge view is provided for handling rebase or merge conflicts.

Technical Details and Availability

Built on libghostty (the engine behind the Ghostty terminal), Farol renders like Ghostty and reads existing Ghostty configurations. The application wrapper is written in native Swift. It is free and MIT licensed, requiring no account and collecting no telemetry. Currently, it supports Apple Silicon devices running macOS 14 or later.

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.