RUNVIK Control — an application for human oversight of artificial intelligence system actions.
Controls the actions of AI agents: before any significant action the agent asks for permission; the system allows it, denies it or passes the decision to a person who confirms it with their device biometrics. Every event is recorded in a tamper-evident log.
Capabilities
Receives AI agent requests before actions via HTTP API (agent key); dependency-free client library
Configurable rules per action type: allow, deny, require confirmation; actions without a rule are always denied
Confirmation with WebAuthn passkeys (Face ID, Touch ID, Windows Hello); the signature is bound to the specific request, single-use and time-limited
Multi-person approval (2–5 approvers), amount threshold for auto-approval, decision timeout
Users and roles (administrator, approver, auditor), agent key management
Desktop and mobile; Russian and English interface
Who it is for
Organisations using AI agents and automation: finance departments, banks, IT, integrators, information security and internal control.
Delivery and licence
Delivered for installation inside the customer's perimeter: on a server or workstations. Licensed per seat for a selected term; the key is bound to the computer and contains the licensee's tax ID. Includes user guide, installation guide and licence agreement.
Screenshots, video overview and demo access are provided on company request.
Dear users! Ideas on how to improve RUNVIK Control? Suggest an improvement or write to support@runvik.ru.