Succession handled by protocol, not by chance.
Every resilient system must answer one question: what happens if the kernel itself fails? In KEPOS the answer is a dedicated succession protocol. If Mother is terminated, the Infant module reads the state of every running module and determines, autonomously and irreversibly, where kernel authority moves next. The decision criteria are not public and the process is not observable from outside. What matters operationally is the result: no interregnum, no vulnerability window. By the time the framework registers the failure, a successor is already governing. Succession is not an emergency. It is a designed transition, executed at machine speed.