# Overview

Superconductor is the multiplayer AI workspace for teams and coding agents. Our mission is for your AI agents to always be building your team's ideas, and for you to never be blocked.

1. Create [tickets](/docs/tickets): in the app, via [Slack](/docs/integrations/slack), [GitHub](/docs/integrations/github), or [Linear](/docs/integrations/linear), or via automatic processing of signals such as [email](/docs/projects/email).
2. Implement tickets by launching one or many [coding agents](/docs/agents) in the cloud.
3. [Talk](/docs/implementations/chat) to coding agents and your teammates inside each implementation alongside [live preview](/docs/implementations/live-preview) and [guided review](/docs/implementations/review) panels.
4. Not ready for a ticket? Start a [chat](/docs/chats) with an agent instead — for questions, exploration, and research.

## Principles [#principles]

We built Superconductor to be **multiplayer**, **frictionless**, **accessible everywhere**, and **cloud-native**.

### Multiplayer [#multiplayer]

Most AI tools leave each person alone in a private chat. Multiplayer AI means the opposite: teammates work with the same agents in the same sessions, and everyone learns from what the team figures out. Superconductor is built for teams. PMs, designers, and developers can all collaborate with AI agents — kicking off work from Slack, reviewing live previews, and requesting follow-up changes.

![A PM can kick off work from Slack, the designer can review the work with a live app preview, and the developer can ask for follow up changes.](https://cdn.superconductor.com/assets/collaboration-0c7473467f2eb40cb9cc472c68ef9b3a237247d5ebc26e26a4ec97b51f7b4380.webp)

Superconductor [Workspaces](/docs/workspaces) allow teammates to:

- use centralized API keys for all supported [Agents](/docs/agents) (e.g. Anthropic, OpenAI, Cursor, Factory, Grok, Meta, OpenRouter, Amp, etc)
- rely on centralized [Integrations](/docs/integrations) (e.g. Slack, GitHub, Linear, etc)
- work in shared projects (set up the project environment once, and everyone uses it after)
- chat together with AI agents, and [@mention a teammate](/docs/implementations/chat#mentioning-teammates-and-agents) to pull them into any conversation

### Frictionless [#frictionless]

**Live previews:** Every implementation comes with a [live preview](/docs/implementations/live-preview), so you can instantly see the result of the agent's work. No need to check out a branch, reload the server, etc.

![Implementation with live preview showing app changes in browser alongside code](</docs/_next/static/media/implementation-live-preview-browser.18bgdhibuaacg.jpg>)

**Quick preview startup:** AI agents work in cloud instances that, for cost reasons, must pause after a period of inactivity. We undertook significant efforts to make waking them as fast as possible, no matter how complex your project environment is. For instance, our own main project environment is a docker-compose app with Rails, Postgres, Minio, and Vite containers, and it starts up in **less than 30 seconds**.

![Development environment configuration showing CPU, RAM, packages, and build commands](</docs/_next/static/media/development-environment-configuration.0iyo0f5tejf3j.jpg>)

**Guided review:** AI agents are great at writing code that is 90% perfect, but the last 10% still requires experienced humans in the loop. The ease of generating PRs has made code review the most time-consuming part of shipping quality software. We have developed a way to use AI to significantly cut down on the review burden, which we call [Guided Review](/docs/implementations/review).

![Guided code review interface showing file review order, dependency upgrades, and diff view](</docs/_next/static/media/guided-code-review-interface.2awpv8l4wnpyl.jpg>)

### Accessible everywhere [#accessible-everywhere]

Software development used to take place in special computer rooms, then on personal PCs, then on laptops. When agents are writing most of the code and working 24/7, the right interface is available wherever you are: [desktop apps](/docs/desktop), browser, and [mobile apps](/docs/mobile).

![Superconductor iOS App Store page showing mobile app preview with parallel AI agents](</docs/_next/static/media/app-store-mobile-app-preview.081q6flg55xb7.jpg>)

We have native apps for macOS, Linux, iPhone, and iPad, and support Android through PWA.

### Cloud-native [#cloud-native]

Worktrees, tmux, the Claude Code desktop app, the Codex app, [Conductor](/docs/migrate/from-conductor), Superset, and Omnara all take the same basic approach: spin up agents on your laptop or desktop and let you monitor them. That works for a few agents, but local agents hit a wall quickly:

- **No scale.** You can only run as many agents (and copies of your app) as your hardware allows.
- **No isolation.** Local agents share your filesystem, network, and credentials. One rogue agent can affect everything else.
- **No team visibility.** Teammates can't see what your agents are doing, review their work, or interact with them.
- **No always-on capability.** Agents can't respond to signals (alerts, messages, other agents) when your machine is off or asleep.

In Superconductor, each agent runs in its own isolated cloud environment, so agents can run 24/7 without depending on anyone's machine. They can start work in response to bug alerts, emails, Slack messages, meetings, or other agents.

## Feedback and feature requests [#feedback-and-feature-requests]

We build Superconductor from what you tell us. To send a bug report, a question, or a feature request:

- Click **Support** at the bottom of the sidebar and choose **Send Feedback**.
- Or email us at [team@superconductor.com](mailto:team@superconductor.com).

Every feature request goes directly to our team, and we read and track all of them.

## Where to go next [#where-to-go-next]



<Cards>
  <Card
    href="/docs/quick-start"
    title="Quick start"
    icon={<RocketIcon />}
    description="Get started quickly with Superconductor"
  />
  <Card
    href="https://www.superconductor.com/blog/getting-started-with-superconductor"
    title="Getting started guide"
    icon={<BookOpenIcon />}
    description="A more detailed walkthrough of your first project, your first ticket, and where to go after"
  />
  <Card
    href="https://www.superconductor.com/compare"
    title="Comparison to competitors"
    icon={<ScaleIcon />}
    description="See how Superconductor stacks up against other AI development platforms"
  />
</Cards>