Glossary · always-on-agent
Always-on agent
Definition
An AI coding agent that runs continuously rather than starting and stopping with a developer's session. Always-on agents handle background work (overnight refactors, long test runs, multi-step migrations, dependency upgrades) and notify the human only when an action requires approval. They live on a server, not a laptop.
Context
An always-on agent is the operational cousin of a long-running agent: long-running describes a single task spanning many hours; always-on describes the agent's lifecycle, in which sessions don't really begin and end. The user opens a chat, gives direction, closes the laptop, walks away, and reopens later from a phone or another device. The agent has been working the whole time. Approval requests buffer in a notification queue and resolve when the human is next reachable.
Also known as
Related terms
See also