Managed agent runs
A managed agent run is a coding-agent task executed under management: scheduled or triggered by the team, run on the team's own machine or CI, bounded by a hard token and time budget, and reported back with its status, cost, and a verifiable result.
The difference between running an agent and a managed run is accountability. A managed run has an owner (the ticket's author), a fire-point (run at 2am, or on a trigger), a budget the runner enforces live by killing the process when it is exceeded, and a report the operator reads the next morning.
In Aethereum a managed run starts as a ticket marked auto with a runAt time. The scheduler queues it exactly once; a machine that has explicitly opted in executes it under guarded autopilot; the run's status, token count, and cost stream to the room while the full transcript stays on the machine that ran it.
[schedule your first managed run →]
related terms