Run Claude Code with your whole team, in your own cloud
Published July 19, 2026 · by the AQ team
AQ runs Claude Code, Anthropic's agentic coding CLI, in cloud workspaces your whole team shares: the same live terminal, code editor, and app preview, streamed to everyone's browser. Sessions run in tmux on a VM you control, so they survive closed laptops and lost Wi-Fi, and each workspace gets its own isolated git worktree, so parallel runs never collide. You sign in with the Claude subscription you already have; AQ never proxies or marks up your model usage.
Claude Code itself is excellent single-player. What AQ adds is everything around it that starts to matter the moment more than one person cares about the work: persistence, parallelism, visibility, and handoff.
What running Claude Code in AQ adds
- Always-on sessions. Claude Code runs in tmux on the workspace VM, not on your laptop. Close the lid, switch devices, or lose your connection: the run keeps going, and the full scrollback is there when you come back. (The long version: keep Claude Code running after closing your laptop.)
- Parallel sessions without collisions. Every workspace is an isolated git worktree on the same repository, so five Claude Code sessions in five workspaces work five tasks with zero conflicts. (How and why: run multiple Claude Code sessions in parallel and git worktrees for AI coding agents.)
- Multiplayer, not screen-sharing. A teammate opens your workspace URL and is in the actual session: same terminal, same editor, same running preview. They can watch Claude Code work, type into the session, or take over entirely. (Details: share a Claude Code session with your team.)
- Previews people can react to. Each workspace can serve its dev server with a shareable link, including guest links for people outside the team, and comments pinned on the preview can be sent to Claude Code as prompts.
- Work arrives from your tracker. Label a Linear issue "ai-task" and a workspace appears in the team sidebar, with ownership following the assignee and status syncing back.
- PRs tracked per workspace. Claude Code commits, pushes, and opens pull requests with your own GitHub sign-in, and AQ records opened, merged, and closed PRs against the workspace.
How setup works
Three steps, a few minutes total. First, connect Claude in AQ's settings: it is the standard interactive Claude Code sign-in, done once per user, using whatever Claude plan you already pay for. Second, connect a repository and choose where agents run: a VM you bring on the Free plan, or a dedicated AQ-managed VM on the Team plan. Third, create a workspace (from a prompt, a branch, or a Linear issue) and start Claude Code from the workspace's terminal. Teammates who open the workspace join the same session; there is nothing extra to install for them.
What AQ deliberately does not change
AQ is a harness, not a fork. The Claude Code inside an AQ workspace is the unmodified CLI with your own login and your own settings, running in a real terminal. Anything Claude Code can do in a terminal, it can do in AQ, and anything you can type locally, you and your teammates can type here. If Anthropic ships a new Claude Code feature, you have it the moment the CLI updates, not when AQ catches up. And if you leave AQ, your workflow ports out cleanly: it was always just Claude Code, git worktrees, and tmux.
Claude Code alongside every other agent
AQ is agent-agnostic: the same workspace model runs Codex, Cursor Agent, Kimi, Grok, Pi, OpenCode, and plain shells. Teams use that for side-by-side comparisons (same task, two CLIs, two worktrees) and for mixing strengths, like Claude Code for a deep multi-file change while Codex handles a mechanical migration next door. See running Codex with your team for the same story from the other side.
Frequently asked questions
Do I need an Anthropic API key to run Claude Code in AQ?
No. You sign in with your existing Claude subscription through the normal Claude Code login flow, once per user. AQ does not sit between the CLI and Anthropic, and it never marks up or resells model usage.
Does Claude Code cost extra inside AQ?
Your Claude usage is billed by Anthropic exactly as it would be locally. AQ itself has a Free plan (bring your own VM) and a Team plan at $100 per user per month promotional (standard $200) that includes a dedicated AQ-managed, always-on VM.
Can several people use one Claude Code session at the same time?
Yes. An AQ workspace streams the same tmux session to everyone who has access: all of them see the same terminal live, and any of them can type, interrupt, or take over. That is the core difference from single-player tools where a session belongs to one machine.
How many Claude Code sessions can I run in parallel?
As many workspaces as your VM can handle: each workspace has its own git worktree, so running one session per workspace means no collisions on files or branches. Teams routinely run several agents against the same repository at once this way.
Does this work from Windows or Linux?
Yes. AQ runs in the browser, so the full Claude Code experience (terminal, editor, preview) works the same from macOS, Windows, Linux, or a phone. The CLI itself executes on the workspace VM, not on your device.
Is Claude Code in AQ modified in any way?
No. It is the standard CLI running in a real terminal with your own login. AQ adds the workspace around it: isolation, persistence, sharing, previews, and tracker intake.