Nous Research's Hermes can repeatedly check a changing situation through its recurring loop feature. Each wakeup runs a fresh AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → turn against the current state, such as the latest build result, then waits for the next interval. A loop can stop after a run count, a completion condition or a user command. It belongs to a conversation and requires its backend process to remain running; the documentation directs work that must survive terminal restarts to scheduled cron jobs. This makes loops useful for monitoring during an ongoing work session.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
