Markov Decision Process

An MDP formalizes RL with states, actions, transitions, and rewards under the Markov property.

What key assumption does a Markov Decision Process (MDP) make?