Reading your audit trail

Every autonomous action is logged, explained, and — where possible — reversible. How to read the timeline, export it, and why it matters for disputes and compliance.

The audit log is a single chronological record of every agent action and every policy change on your account. Each entry shows who acted (the agent, you, or the system), what they did, when, the entity it touched, and the outcome. You can filter the timeline by actor and by date range to narrow in on a period or a decision.

Alongside it, the agent activity feed gives the plain-language "why" for each autonomous decision — the reasoning the agent used, not just the result — so a logged action is always explainable after the fact, not a black box.

Entries carry a reversibility tier, so you can tell at a glance whether an action can be undone. Reversible actions can be walked back directly from the activity feed with an undo control; actions still inside their soft-commit window can be cancelled before they ever execute. The two together — "you can always see it" and "you can always take it back" — are the point of the trail.

Because the log records the policy in force at the moment of each action, you can also see the exact thresholds the agent was operating under when it made a given decision, not just the settings you have today.

The log is tamper-evident: entries are hash-chained, so any later alteration or gap in the record is detectable rather than silent. On the web app you can export the filtered timeline to CSV and produce a Decision Integrity Certificate that attests the chain has not been altered.

This is what makes the trail useful outside the app. If a tenant disputes a charge, a fee, or a notice, or if a regulator or your attorney asks how an automated decision was made, you can hand over a dated, verifiable record of exactly what happened and under what policy — rather than reconstructing it from memory.