In this paper, we present a Symbolic Reinforcement Learning (SRL) basedarchitecture for safety control of Radio Access Network (RAN) applications. Inparticular, we provide a purely automated procedure in which a user can specifyhigh-level logical safety specifications for a given cellular network topologyin order for the latter to execute optimal safe performance which is measuredthrough certain Key Performance Indicators (KPIs). The network consists of aset of fixed Base Stations (BS) which are equipped with antennas, which one cancontrol by adjusting their vertical tilt angle. The aforementioned process iscalled Remote Electrical Tilt (RET) optimization. Recent research has focusedon performing this RET optimization by employing Reinforcement Learning (RL)strategies due to the fact that they have self-learning capabilities to adaptin uncertain environments. The term safety refers to particular constraintsbounds of the network KPIs in order to guarantee that when the algorithms aredeployed in a live network, the performance is maintained. In our proposedarchitecture the safety is ensured through model-checking techniques overcombined discrete system models (automata) that are abstracted through thelearning process. We introduce a user interface (UI) developed to help a userset intent specifications to the system, and inspect the difference in agentproposed actions, and those that are allowed and blocked according to thesafety specification.