Fleet · org-wide control
Fleet: every agent. One command center.
Every agent on every machine you connect — one live board, one set of controls.
Beta · Included on Pro, Team and Enterprise (not on Free)
Nothing runs unseen
An agent you can’t see is an agent you can’t trust.
Every machine, one map
Laptops and runners register themselves the moment an agent starts.
Runs stream in real time
Steps, edits, and test results stream in — no terminals to tail.
The whole org, not just you
Every team’s fleet in one governed view, not scattered sessions.
A hand on the wheel
Every run carries live controls, and agents can stop to ask you a question.
Pause & steer
Stop any run in seconds, edit the ticket, and relaunch on the same machine. The worktree and output are preserved.
Agents ask when unsure
A run can park itself with a question and wait for your reply, and you can stop any run in seconds.
Schedule the fleet
Nightly audits, weekly reports — on any machine.
Everything on the record
Agent requests to the API and MCP land on the append-only audit trail.
Fleet, explained
Fleet is one live board of the machines you connect and the agent runs on them, with pause, stop, approvals and schedules. Machines and runs from your whole organization appear in one governed view.
Fleet is in Beta and is included on Pro, Team and Enterprise. It is not part of the Free plan.
Fleet runs agents on machines you connect, using your own repository access, so source code and working files stay in your environment. On Enterprise you can also route model inference to your own provider endpoint with your own key.
A schedule launches agent work onto the fleet at a time you choose, once or on a repeating schedule. Each repeating fire creates a new task from a template and launches that, so every run has its own claim, comments and pull request.
Fleet is in Beta.
Fleet is in Beta on Pro and above
Fleet runners are included on Pro, Team and Enterprise and are not part of the Free plan. See the plans to find the right fit.