governance
AI governance / AI Act-oriented architecture review
Dry review direction for architecture, feasibility, and public evidence boundaries before governance language drifts into abstraction.
Addresses: the gap between governance claims and the actual architecture / release / evidence contour.
Primary supporting pages: AGI, L4, Evidence, Library.
Not this: legal advice, instant certification, or procurement shortcut.
auditability
L4 Witness / evidence-trail / auditability review
Review direction for witness trail, auditable privileges, release boundaries, and fail-closed operational design under real constraints.
Addresses: how reviewability stays legible when identity, privilege, and evidence surfaces must survive contact with reality.
Primary supporting pages: L4, Evidence, Releases, Downloads.
Not this: forensic guarantee, managed SOC service, or compliance badge generator.
local-first
Sovereign / confidential / local-first AI architecture review
Review direction for continuity, sovereignty, confidential boundaries, and local-first implementation without collapsing the discussion into generic tooling.
Addresses: the boundary between long-lived entity framing, confidential/local-first design, and executable layers.
Primary supporting pages: SER, AGI, Publications, Library.
Not this: managed infrastructure outsourcing, generic chatbot integration, or cloud migration sales copy.