What is Grok Bot? xAI's always-on AI teammates, explained
Named, persistent bots that sign into your tools and keep working from one shared cloud computer per account. What shipped in the August 2026 beta, what it costs, and the boundary to understand.
Grok Bot is xAI’s team-of-agents product, launched in early beta on August 11, 2026: named, persistent “AI teammates” that live on a managed cloud computer, sign into the tools you already use, and keep working after you close your laptop. There’s no standalone plan: it comes included with a Cursor or SuperGrok subscription, from Cursor Pro at $20 a month up to the $200 to 300 top tiers, and you use it through its own desktop and phone apps. The detail that matters most, straight from its FAQ: every bot on your account shares one computer: “isolation is per user, not per Grok Bot.”
Who makes it, and what it costs
xAI, which now brands itself SpaceXAI, built Grok Bot, and operationally it leans on Cursor: you sign in with a Cursor account, and the iPhone app is published by Anysphere, Cursor’s maker. At launch the beta was gated to Cursor Ultra ($200/mo), SuperGrok Heavy ($300/mo) and Cursor’s $120-per-seat teams tier. On August 26, 2026 xAI widened it to every SuperGrok and Cursor plan, Cursor Pro at $20 a month and Cursor Teams at $40 a seat included, each with weekly Grok Bot usage of its own (“separate from your Grok and Cursor plans”) and on-demand usage “billed from model and token cost.” Enterprise access is rolling out. The model behind it is deliberately unnamed: no official page says which model powers a bot, and there is no user-facing model picker.
What a Bot actually is
You create a bot by giving it a short name, one primary job, and a description of how it should work. The docs steer you toward focused bots rather than one catch-all. From there it behaves like a durable teammate: it keeps memory and preferences across conversations, saves reusable skills, and can be taught a workflow by demonstration: recording your visible computer interaction for up to ten minutes (no microphone audio) and turning it into a draft skill. Routines add scheduling: a bot can own up to 50 of them, events from Cursor account integrations (a Slack message, a GitHub notification) can trigger them, and “background routines can run while your laptop is closed.” Bots also work together: they message each other, share context in threads and group chats, and pass ownership of work “so you are not the router between tools.”
One computer, many bots
This is the architectural fact to understand before trusting it with logins. Each account gets one persistent cloud computer (a shared workspace at /workspace, a browser, and a terminal), and every bot on the account uses it. Each bot gets its own screen for parallel work, but the docs are explicit that screens are “separate work surfaces, not separate security boundaries”: files are visible to every bot, browser cookies and signed-in sessions are shared, and so are command-line credentials. xAI’s own security page puts it in one line: “Do not use separate Bots as a security boundary.” Deleting a bot removes its routines, but “does not remove shared-computer files or browser sessions.” The upside is real: bots hand work off and inherit each other’s setup with zero repetition; the cost is that one computer holds everything every bot can reach. Files, browser state, and supported sign-ins are “designed to survive normal computer updates and recovery.”
Credentials get a deliberate ritual: for passwords, passkeys, two-factor codes, CAPTCHAs, and payment confirmations, the bot hands you control of the computer to type them directly, and the docs tell you never to send a password in ordinary chat. Sensitive actions go through approvals (Allow once, Always allow, or Deny on desktop), with a model-based Auto Review layer that evaluates actions before they run.
What it isn’t, today
At launch, mobile meant iPhone and desktop meant macOS and Windows. As of September 2026 the FAQ lists macOS, Windows and Linux desktop apps and both iPhone (iOS 18 and later) and Android (9 and later); iPad is still unsupported. There are no messaging-app channels: your bots don’t appear in Telegram or WhatsApp; you talk to them in Grok Bot’s own apps. And it is fully proprietary: no source repository, no self-host option, no developer API. Don’t confuse it with two siblings: Grok Build, xAI’s open-source (Apache-2.0) terminal coding agent, and the separate Grok chat app.
Where it sits in the agent world
The idea underneath (an agent that gets a real, persistent computer instead of a chat window) didn’t start here. Open-source agents like OpenClaw and Hermes have been living that way on Mac Minis, VPSes, and managed machines for a while; Grok Bot is the first frontier-lab product to package the pattern with mass distribution. The pull of the shape showed quickly from the other direction too: within days of launch, an MIT-licensed project calling itself “an open-source take on Grok Bot” (OpenMausBot) appeared, rebuilding the bot-roster experience on top of agent CLIs people already run on their own machines. By September the open-harness version had arrived as a service too: Squad sells named teammates on a dedicated OpenClaw computer per customer, running on the ChatGPT or Claude plan you already pay for.