Poly vs AQ for running coding agents as a team

AQ and Poly both let a team join one live coding-agent session, and they are built for different people. AQ runs real agent CLIs such as Claude Code and Codex on a dedicated machine, in your own cloud if you want, and teammates join the same live terminal, editor and app preview. Poly (usepoly.co) runs one AI, Claude or ChatGPT, in a browser room where every member can prompt, approve each change before it happens, and undo any turn, with a receipt on each. AQ is for engineering teams who already live in the CLI. Poly is for small groups where at least one person does not.

What AQ does

AQ's pitch is "run every coding agent in the cloud, with your team." A workspace is an isolated git worktree on a dedicated VM; Claude Code, Codex, Cursor Agent, Devin and others run there under each user's own CLI login. Teammates open the session by link and see the same terminal, editor and running app; reviewers can pin comments on elements of the app and send them to the agent as prompts. Label a Linear issue and AQ opens a workspace and a pull request. The Team plan puts the VM in your own cloud account. The free plan is a personal sandbox that sleeps when idle; Team is $50 per user per month on promotion, rising to $100, listed at $200.

Where AQ is stronger

• Real terminals, real CLIs, your own cloud. If your team's workflow is Claude Code in a terminal, AQ is that terminal, shared, with full custody of the machine.
• Engineering integrations. Linear issues to pull requests, a REST API, webhooks, an MCP server.
• Any agent. Six or more CLIs, each on the user's own subscription with no markup.
• Comment on the running app. Pinned comments that become prompts is a good idea, and AQ shipped it.

Where Poly is stronger

• A say before it acts. AQ's only approval is a control request to type in someone else's terminal. Poly's approval modes put every proposed file change and command on every member's screen in plain language, before it happens.
• Receipts. Every Poly turn names who ran it, which model, how long and what it cost, charged to that person. AQ bills per seat and the agent spend is on each user's own subscription.
• No keys, no CLI logins. Nobody in a Poly room configures anything. AQ users sign into each CLI themselves.
• Built for the non-engineer. AQ names PMs and designers, but the surface is a terminal. Poly's surface is a timeline, cards and a preview.
• A channel the AI cannot read. Poly's group chat and polls are hidden from the agent.

Pricing, side by side

AQ: free personal sandbox; Team $50 per user per month for the first year on promotion, then $100, list $200. Poly: free to start with no card; $20 or $50 a month for each person who runs the AI, and everyone else watches, chats and approves for free. A five-person team with two people prompting pays Poly $40 and AQ $250 on promotion.

Which should you pick?

• Engineers who want a shared cloud terminal for the CLIs they already use, with Linear and GitHub wired in: AQ.
• A team that wants to keep the agent's machine in its own cloud account: AQ Team.
• A small group including people who do not code, who want to approve each change and see who spent what: Poly.
• You want ChatGPT and Claude in one room without anyone holding a subscription: Poly.