How the ops agent works
Observe, predict, remediate, optimise.
The ops agent reads metrics, logs, traces and deploy events for every resource in a cloudspace and joins them into one timeline per service.
The loop
- Observe — collect signals and learn each service’s normal baseline.
- Predict — flag trends that will breach a limit, such as memory growth or disk filling up.
- Remediate — take an allowed action: scale, restart, roll back or rebalance.
- Report — post what happened, why, and the result to your chosen channel.
Every action is recorded in the cloudspace audit log with the evidence the agent used.