What the work covers
Human approval points, decision rights, consent and attestation records, audit trails, architecture decision records, source-of-truth documentation that agents must read before they act.
Single sign-on, multi-factor and step-up authentication, role-based access, tamper-evident audit logs, guarded destructive actions, threat modeling.
Human control by design
Agents prepare the work and a named person approves each decision before it takes effect. Who approves what, and what is logged, is settled before anything is built.
Security built into how we work
Security is a property of how Resolon builds, not a separate service. Work happens inside the client's own environment, access is limited by role, and every action is logged.
3Rules before anything is built
Who approves what, what gets logged, where AI may not act, and the policy for how it is used: settled and written down first.
Where it sits: roadmap stage 3