explaingit

o3dotdev/o3-code

Analysis updated 2026-05-18

28TypeScriptAudience · developerComplexity · 3/5Setup · moderate

TLDR

A macOS app that runs multiple AI coding agents at once, each in its own isolated Git worktree, from a single dashboard.

Mindmap

mindmap
  root((Superset))
    What it does
      Runs multiple AI coding agents
      Worktree isolation per task
      Single dashboard view
    Tech stack
      TypeScript
      Bun
      Git worktrees
      Docker
    Use cases
      Parallelize AI coding tasks
      Compare agent outputs
      Review diffs before merging
    Audience
      Developers
    Platform
      macOS only
      Windows untested
      Linux untested

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Run several AI coding agents like Claude Code and Codex on different tasks at the same time.

USE CASE 2

Inspect what each agent changed with a built-in diff viewer before accepting the changes.

USE CASE 3

Hand off any agent's workspace to your usual code editor or terminal with one click.

USE CASE 4

Automatically run setup and teardown scripts for each new agent workspace.

What is it built with?

TypeScriptBunGitDocker

How does it compare?

o3dotdev/o3-codeaudemodo/audemodo-responsive-keepalivecahn91/gpt-repo-mcp
Stars282828
LanguageTypeScriptTypeScriptTypeScript
Setup difficultymoderateeasymoderate
Complexity3/52/53/5
Audiencedeveloperdevelopervibe coder

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

macOS only currently, requires Git 2.20 or later and Docker for local development.

In plain English

This repository contains the source code for Superset, a macOS desktop application that lets you run multiple AI coding agents at the same time from a single dashboard. AI coding agents are programs like Claude Code, OpenAI Codex, Cursor, and Gemini CLI that can read and write code in a terminal. Normally you can only work with one at a time, and each agent needs its own working copy of your project. Superset manages that juggling for you. The core idea is worktree isolation. When you give Superset a task, it creates a separate working directory for that task by using a Git feature called worktrees, which lets multiple branches of a project coexist on disk simultaneously. Each agent gets its own isolated copy, so several agents can work on different parts of a project at once without interfering with each other. You monitor all of them from one place. The interface includes a built-in diff viewer so you can inspect what each agent changed without opening another application, and one-click handoff to open any workspace in your usual code editor or terminal. You can configure setup and teardown scripts that run automatically when a workspace is created or deleted, which handles things like copying environment files or installing dependencies for each agent task. Superset works with any command-line coding agent, not just a fixed list. The README lists Claude Code, OpenAI Codex, Cursor, GitHub Copilot, Gemini CLI, and others as fully supported, and notes that anything runnable in a terminal should work without extra configuration. The app is currently macOS only, Windows and Linux builds are listed as untested. It is built with TypeScript using the Bun runtime and requires Git 2.20 or later. The source code is open and includes a local development setup that uses Docker for a local database.

Copy-paste prompts

Prompt 1
Help me install Superset and set up my first AI coding agent workspace.
Prompt 2
Explain how Superset uses Git worktrees to isolate multiple agent tasks.
Prompt 3
Walk me through configuring a setup script that copies my environment files for each new workspace.
Prompt 4
How do I add support for a new command-line coding agent in Superset?

Frequently asked questions

What is o3-code?

A macOS app that runs multiple AI coding agents at once, each in its own isolated Git worktree, from a single dashboard.

What language is o3-code written in?

Mainly TypeScript. The stack also includes TypeScript, Bun, Git.

How hard is o3-code to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is o3-code for?

Mainly developer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.