Agentic Modules
The Auditor is a digital employee for quality. Every 15 minutes it sweeps for outliers and drift. Inside the authority you set, it opens the hold.
How it works
The sweep follows the same fixed rules every time, so the same record always gives the same flags. AI is used for one thing only: drafting the root cause for you to edit.
Sweep log
Every 15 minutes, the Auditor checks each quality metric against its own baseline and flags anything that sits too far outside it. Every finding gets a severity, from low to critical, so you know what needs attention first.
Field drift
It tracks every numeric field on your submitted forms over time, including ones you might not think to watch. Eight readings in a row on one side of centre is enough to flag drift. The finding names the field, the form and the baseline it drifted from.
Root cause
The root cause draft cites the readings and records it was built from, so you can see the working behind it. Similar closed cases come in as context, the ones your team already solved. You edit the draft, add the countermeasure, and close the case yourself.
Hold rules
You set whether a hold runs on its own or waits for you, per action, in the Authority Matrix. A routed hold sits in the Approval Queue until you decide. Your decision wakes the Auditor back up, and it opens or drops the hold.
Plan events
When a hold opens, the Auditor publishes it, along with every anomaly, resolution and drift finding it makes. The Planner picks that up and replans the jobs the held lot was feeding. Notifications reach the right people by role and by site.
Ask
Ask the Auditor why a lot is held, or what closed the last one, and it answers from the case history. Every run records its own metrics, visible in the admin tabs. Two golden sets check its answers against known outcomes.
Finding rules
A drift finding raises an event and a notification, and that’s where it stops: no root cause draft, no proposed hold. Findings on your quality metrics are different; those carry the full path, from draft to hold. Know which kind you’re looking at before you act on it.
Every reading gets checked against its own baseline, sweep after sweep. When eight readings in a row land on the same side of centre, the finding names the field and the baseline it drifted from.
When a quality metric is out, you get a root cause draft built from the quality record and the cases that closed before it. You edit it, add the countermeasure, and close the case.
Only where your authority rules say it can. Everything else waits in the Approval Queue, and the decision there wakes it to finish the job.
Every 15 minutes, per site, on a subscription the platform keeps running. A governance decision also wakes it between sweeps.
Statistical outliers on quality metrics, and drift runs on numeric form fields. Each finding carries the baseline it was measured against and a severity.
The hold is published, and the Planner replans the jobs the held lot was feeding. The people who need to know get a notification.
30 minutes on your record and what the Auditor would flag before the lot moves.
We use cookies to understand how the site is used and to connect form submissions to earlier visits. No analytics or marketing cookies are set until you choose. Privacy Policy