super.engineering

super.engineering is a native macOS app for running parallel AI coding agents in isolated git worktrees. Review diffs, manage branches, and coordinate sessions from one keyboard-driven workspace.

AI Agent Development AI Developer Tools Programming Agent

Visit Website

Overview

super.engineering is a native macOS app for running parallel AI coding agents in one place. It is built in Rust and uses isolated git worktrees so you can launch multiple CLI agents, review their output, and manage the resulting branches without switching tools.

The product is aimed at developers who want a keyboard-driven workspace for agentic development. It brings together agent orchestration, diffs, PR tracking, terminal sessions, layouts, themes, and repo-level automation in a single native interface.

Features

Repository-aware setup

Open any git repository, let the app detect configured agents, and start from a ready workspace.

Parallel agent spawning

Create isolated git worktrees and launch agents at the same time so each session stays separated.

Flexible workspace layouts

Switch between stacked, horizontal, or bottom terminal arrangements and float tabs in separate windows.

Git workflow controls

Review diffs, track PR status, manage branches, and use a smart action button for commit, push, PR, conflict resolution, and merge steps.

Terminal and chat views

Toggle between raw terminal output and a richer chat view for the same session.

Automation and custom commands

Define reusable prompts, scripts, and per-repository setup or teardown steps, then scope them globally, per workspace, or per project.

Use Cases

Parallel coding across one repo
Open a repository, detect configured agents, and start several sessions in separate worktrees when you want to compare approaches or split work across tasks.
Cross-repo coordination
Keep frontend and backend repositories in one workspace, with shared branches and a single place to review changes across codebases.
Agent review and shipping
Use the terminal, chat view, diffs, and branch controls together when you need to inspect output, review changes, and merge from the same interface.
Repeatable team workflows
Set up repository-specific scripts, prompts, and teardown steps so a team can standardize how sessions start and finish.
Multi-window monitoring
Float a tab, switch layouts, and keep active sessions visible while working elsewhere on the machine.

Pros and Cons

Pros

Runs as a native Rust app rather than an Electron-based web wrapper.
Supports multiple CLI agents in isolated git worktrees.
Keeps terminal, chat, branches, diffs, and PR status in one interface.
Offers configurable layouts, themes, fonts, and floating tabs.
Supports per-repository scripts and reusable custom commands.
The Teams page describes a private-by-default workflow for organizations.

Cons

The source does not provide a public pricing table or clear plan breakdown.
Platform support appears macOS-focused in the available copy.

FAQ

What is super.engineering?

super.engineering is a native macOS app for running parallel AI coding agents. It launches agents in isolated git worktrees and provides a single interface for reviewing diffs, branches, and PR status.

How does it manage multiple coding agents?

The app is designed to run multiple agents in parallel from one keyboard-driven workspace. It supports agents such as Claude Code, Codex, Gemini CLI, and other CLI agents, with sessions kept in isolated worktrees.

Is super.engineering suitable for teams?

Yes. The Teams page says super.engineering can be brought into engineering teams as a private native workspace, and that setup help is available for workspace defaults, agent conventions, editor handoff, and rollout paths.

Does super.engineering see our code or prompts?

The Teams page states that agents run as local subprocesses with your existing subscriptions and credentials, and that model traffic goes directly to the providers you already use. The app does not proxy model traffic or store project data.

How is enterprise pricing handled?

The source does not show a public pricing table. The Teams page says enterprise pricing is scoped around team size, rollout needs, and support expectations, and invites you to email the team.