Every action the policy gate flagged for human review. Decision is signed back to the broker; the audit row records who approved, when, and why.

Who asked, who decided, what, when, and under which mode. Signed, tamper-evident export coming.

The mode that decides whether an automated action runs. Out-of-band from the AI — the AI cannot promote itself out of LOCKDOWN. A mode change propagates to every protected system inside ~5s.

View only. In the demo and this web console the modes are shown, not switched: the cards below change nothing.

Per-system YAML defining exactly which roles can call which endpoints. Adding a capability is a code-reviewed change — never a runtime toggle. The audit row records the merge SHA next to every action that capability authorizes.

The AI and automation activity we observe on-prem, minus what already routes through governance. The gap is your ungoverned exposure — highest-signal first.

Every host you declared, and whether its Vertirite agent is reporting in. A host counts as covered only once its own agent sends a heartbeat.

Governed vs ungoverned coverage and your ranked ungoverned exposure — the forwardable summary a CISO, board, or auditor can read. Export it as a PDF anytime.

Your entitlement and the health of Vertirite's own protections — license tier, intelligence freshness, and the self-watch that detects a blocked phone-home or a copied install.

Artifacts you brought under governance from the Coverage map. These are tracked, audited, and policy-gated — no longer part of your ungoverned exposure.

Everyone who asked for the read-only Demo, and where their login stands. Revoke ends a login at once, along with a sign-in link that hasn't been opened yet. It is recorded in the audit log under your name.

Each customer or business unit lives in its own tenant. Per-tenant isolation flows end-to-end: API keys, audit log, capability registry, mode authority — everything scoped.