Resolve

Resolve is CloudThinker’s four-stage incident response flow. Pulse detects actionable problems, Anna investigates them, approved runbooks — pre-written fix procedures — carry out tightly scoped actions, and useful lessons can inform later investigations.

​

How it works

Resolve runs the platform’s agentic loop — Detect → Analyze → Resolve → Validate — end to end; see How CloudThinker works for the loop itself. These are the Resolve-specific stages:

1. Pulse detection — Pulse takes in raw events, suppresses known noise, groups related events into Clusters, and routes actionable ones into Incidents.
2. Parallel RCA — Anna runs a three-phase root-cause analysis (RCA). She can delegate focused, read-only hypothesis checks to temporary subagents — short-lived helper agents — and then records the verdict.
3. Scoped Action — a matching, enabled runbook can become a Scoped Action. Policy decides whether it runs, waits for approval, or is denied.
4. Follow-up and memory — the agent can schedule one follow-up check and write incident memory. Reusable lessons return at the start of a later RCA, not during Pulse classification.

​

Start from the cockpit

Open Resolve → Overview to answer two questions: what needs you now, and what Resolve handled without you. Each attention tile opens the matching slice of the Investigation queue. A source-health banner appears when an enabled polling source — a feed Resolve checks on a schedule — fails or pauses, so a broken intake path does not look like a quiet system.

​

Control automatic investigation

On Duty (the automatic-investigation toggle) in Settings controls whether detection hands off to analysis on its own. Manual investigation always remains available. Turning automatic investigation back on does not start a backlog of Incidents created while it was off.

​

What you can do

​

Get started