Automation & Workflow

Settings → Rules lets admins build automation/alert rules per entity type: a set of AND/OR conditions (including fields on related entities) that, when matched, trigger an action — create a risk record, notify a user, or fire a webhook. Rules can be enabled/disabled, run on a schedule, or triggered manually with "Run now".

Settings → Governance defines validation rules per entity/field — a required-field check or a regex naming convention — each with a severity. "Block" severities stop a save outright if violated; "Warn" severities let it through with a visible warning, giving teams a way to enforce naming/data-quality standards without a hard gate everywhere.

Instead of editing a record directly, a proposed change can be submitted for review with a field-level diff against the current values. Reviewers with write access approve or reject (with notes); the proposer can withdraw it at any time. Conflicting proposals against the same record are flagged. Currently surfaced on Applications and the standalone Change Proposals page.

Decisions, Principles, Solution Designs, Service Blueprints and Engagements can all route through a multi-step approval chain of stakeholders, each with a sign-off deadline; overdue steps trigger a reminder notification. Settings → Delegations lets a stakeholder configure out-of-office routing so their pending approval steps automatically reroute to a delegate for a given date range, with a reason recorded.

On deployments where dual control is enabled, certain destructive admin actions — revoking an API key, removing a legal hold, running an immediate retention purge, or deleting the whole workspace — don't execute immediately. Instead they create a pending request in Settings → Privileged Actions that a *different* admin must approve before it runs, and which expires automatically if nobody acts on it in time.