How it works

The Pool → Scenario → Runner → Sink data flow.

At its core, Tracemill follows a Pool → Scenario → Runner → Sink data flow:

  1. Pools provide reusable data — IP ranges, user lists, CSV datasets
  2. Scenarios define step-by-step event sequences with stateful variables and expressions
  3. Runners execute scenarios, resolving state and generating events
  4. Sinks deliver events to their destination — stdout, Splunk HEC, S3, TCP, or HTTP endpoints

For multi-workload generation, a Job orchestrates multiple scenarios with shared pools and bindings.

From delivery to validation

The CLI stops at delivery: it generates the events and hands them to the sink. Tracemill Cloud continues from there. A job's workloads can declare an expectation — whether a detection should alert on those events or stay silent — and Tracemill Cloud checks it against your SIEM, confirming first that the events were indexed and then that the expected alert fired on them.

Where to go next