The frequency and sophistication of cyber-attacks necessitate robust defense mechanisms to protect critical infrastructure and ensure system resilience. Existing static models often fail to capture the dynamic nature of these threats, leading to ineffective defense strategies. To address this, we propose ThreatResponder, a dynamic adjustment mechanism based on instantaneous state Markov processes. ThreatResponder precisely represents attack behaviors and defense strategies by leveraging real-time data to adjust the number of concurrent attacks the system can handle. Through continuous parameter updates, ThreatResponder predicts attack occurrences and defense effectiveness, enabling adaptive responses to evolving threats. Comprehensive simulation experiments validate ThreatResponder ’s ability to significantly reduce system shutdown probabilities and enhance operational efficiency. The results demonstrate that ThreatResponder accurately models the complex dynamics of cyber attacks and defenses, offering a robust framework for minimizing disruptions and maintaining system integrity. ThreatResponder enhances current defense mechanisms and offers valuable insights for future research in developing more advanced and effective cybersecurity strategies.