This paper addresses the containment control problem for linear multi-agent systems (MASs) under hybrid cyberattacks by designing a memory-based adaptive dynamic event-triggered mechanism. First, to accurately characterize cyberattacks, mathematical models are developed for both Denial-of-Service (DoS) attacks and deception attacks; concurrently, a state observer is constructed to estimate the unmeasurable states of the follower agents. Second, a memory-based adaptive dynamic event-triggered mechanism is proposed by incorporating historical observed states and adaptive updating rules, while considering the impact of cyberattacks. This mechanism proactively integrates the deception attack signal, enabling it to perceive and quantify the impact of malicious data tampering on the system. It also ensures immediate data transmission recovery upon the termination of DoS attacks, thereby substantially mitigating the adverse effects of cyber attacks on system performance. Furthermore, a switched containment control protocol is designed based on cyber defense strategies, utilizing the observed states and the event-triggered mechanism. Through model transformation, a switched error system is constructed, transforming the containment control problem into a mean-square stability problem for switched systems. Then, by constructing a piecewise Lyapunov-Krasovskii functional, stability analysis is conducted to derive sufficient conditions for mean-square exponential stability of the switched system. The observer and controller gains are obtained by solving linear matrix inequalities (LMIs). Finally, simulations are provided to verify the effectiveness of the proposed method.