A local catalog and control plane that lets Codex or Claude search 2,700+ SKILL.md playbooks, pick exact skills, validate a stack and preview an install plan.
johannesjo/parallel-code
Desktop app to run Claude Code, Codex, and other AI coding agents in parallel, each isolated in its own git worktree and branch.
- Stars
- 1,080
- Forks
- 149
- Open issues
- 48
- Last commit
- yesterday
What is it?
Parallel Code is an Electron desktop app for dispatching several AI coding agents at once. When you create a task, it creates a new git branch from your main branch, sets up a git worktree so the agent works in a separate directory, symlinks node_modules and other gitignored directories into that worktree, and then starts the agent there. Supported agents include Claude Code, Codex CLI, Antigravity CLI and Copilot CLI.
Each task keeps its changes isolated, so you can review diffs per task with inline comments that go back to the agent, then merge the branch into main or discard it from the sidebar. A board view groups open tasks by what they need from you, such as waiting for input, working, or ready for a pull request. Tasks also include a shell terminal, a notes panel, and a Markdown canvas rendered as the agent writes.
The app runs on macOS and Linux, is built with TypeScript, Electron and SolidJS, and is MIT licensed. Phone access over Wi-Fi or Tailscale and an Android app are included for monitoring tasks.
Key features
- 01Runs Claude Code, Codex CLI, Antigravity CLI and Copilot CLI from one interface.
- 02Creates a separate git branch and worktree for each task automatically.
- 03Built-in diff viewer with inline review comments sent back to the agent.
- 04Board view sorts open tasks by whether they need input, are working, or are ready to merge.
- 05Keyboard-first controls with configurable shortcuts.
- 06Optional phone access over Wi-Fi or Tailscale.
Use cases
- Trying several agents on the same bug or feature and comparing their diffs.
- Running multiple independent features in parallel without branch conflicts.
- Reviewing agent output per task before merging it into main.
- Monitoring long-running agent sessions from a phone.
Quick start
- 1
Download the latest release for your platform from the releases page
Things to know
- Supports only macOS and Linux, with no Windows support listed.
- The README's getting-started section is cut off, so further install steps are not confirmed.
In collections
Topics
FAQ
- What is parallel-code?
- Desktop app to run Claude Code, Codex, and other AI coding agents in parallel, each isolated in its own git worktree and branch. Parallel Code is an Electron desktop app for dispatching several AI coding agents at once. When you create a task, it creates a new git branch from your main branch, sets up a git worktree so the agent works in a separate directory, symlinks node_modules and other gitignored directories into that worktree, and then starts the agent there. Supported agents include Claude Code, Codex CLI, Antigravity CLI and Copilot CLI.
- Is parallel-code free to use? What license does it have?
- parallel-code is open source under the MIT license. Check the license terms before using it in commercial products.
- Is parallel-code still maintained?
- Actively maintained. The last commit to parallel-code was on Nov 6, 2026.
- What can I use parallel-code for?
- Trying several agents on the same bug or feature and comparing their diffs. Running multiple independent features in parallel without branch conflicts. Reviewing agent output per task before merging it into main. Monitoring long-running agent sessions from a phone.
- How do I install parallel-code?
- Download the latest release for your platform from the releases page
- Where can I get parallel-code?
- The source code and installation instructions are on GitHub: https://github.com/johannesjo/parallel-code
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A multi-platform hook and skill guard for AI coding agents that stops unrequested hashes, checksums, and scope creep in Codex, Claude Code, OpenCode and other hosts.
A self-hosted AI agent team that runs on your machine, using the agent CLI subscriptions you already have to plan, code, review and merge tasks.
A Claude Code plugin with self-correcting memory, persistent FTS5 research wikis, and workflow skills, commands and hooks for AI coding agents.
A lightweight cross-platform IDE for AI coding that runs multiple Claude Code and Codex agent sessions across projects in one window.
Desktop IDE for running Claude Code, Codex, OpenCode and other CLI coding agents in parallel, each in its own Git worktree.