What's New at spektr

Features

Historic risk breakdown

You can now see every risk score calculation a customer has had, not just the latest. Each point in the history shows the score, the category, and the full breakdown behind it as it stood at the time, including when the calculation ran and what triggered it. The history shows whether a change came from the customer's data or from the risk model, and past breakdowns cannot be edited. The same history is available on the customer profile and in cases. [confirm with product: where users open the history view]

What this changes for you: No change to existing workflows. This gives you an audit-ready answer when a rating has moved: what it was, when it changed, and why.

Process link expiry events

Processes with an external, client-facing link now move from pending to expired when the link passes its expiry, instead of showing as pending indefinitely. You can configure a webhook to receive an event when a process expires, and set up monitoring on expiries in an orchestration process. Processes already completed, failed, or otherwise in a final state are not affected. Rolling out now; available on request.

What this changes for you: No change to your current workflows. This becomes relevant once enabled for your workspace, for example to track re-KYC questionnaires that lapse.

Improvements

Connect to another process from an event orchestration outcome

An event orchestration process can now connect to another process as an outcome, so one process can hand off to the next when an event fires. [confirm with product: how the connection is configured]

Website Sentinel timeline and results

The Website Sentinel entry in the process timeline now looks and behaves like every other node. The results table and the PDF export show three fixed indicators for each run: Risk level, Prohibited categories, and Service failure, plus the AI summary of the scan. Individual check results use one vocabulary everywhere: pass, flag, unknown, unchecked. Rolling out behind a feature flag; a scan renders as before until this is enabled for your workspace.

What this changes for you: When enabled, the Website Sentinel results table and PDF export look different: same data, standardised layout. Worth a heads-up to anyone who documents scans from the PDF.

Fewer duplicate cases from monitoring

Case creation no longer reacts to scenario sets that merely repeat or shrink. A re-run that changes nothing leaves the open case untouched, and a genuinely new scenario adds a single trigger to the existing case instead of opening a duplicate. The create-case node can be configured for one case per trigger or one case with all triggers attached, and behaves the same regardless of the workspace-wide alert duplication setting. Rolling out behind a feature flag; available on request.

What this changes for you: When enabled, daily monitoring creates fewer duplicate cases for the same customer.

Alert nodes in event orchestration

You can now add Alert nodes to event orchestration processes. [confirm with product: scope of this one, the source page is empty]

This release also includes bug fixes, platform and performance work.