Read The Black Box

Reading Execution Telemetry

Every AI Hub run leaves a complete audit trail.

Steps

1. Open Admin → AI Hub → Execution sessions
2. Click on a successful session
3. Scroll to Execution step runs
4. Expand a step run and read:
5. Request payload: what was sent to the model
6. Response payload: what the model returned
7. Latency ms: how long the call took
8. Return here and click Check Mission

Why this matters

Full telemetry at every step lets you: - Debug unexpected outputs - Audit AI decisions for compliance - Measure cost and latency per step - Reproduce exact conditions for a run