The great proliferation of social media applications has ignited serious concern about the privacy of users, the safety of their data, and susceptibility to phishing attacks. Among the new threats that had arisen, unauthorized screenshots and phishing activities turned out to be those that could peak in different proportions and harm in one way or another. The system proposes a composite design for detecting and preventing screenshots within social media environments, using a combination of methods to combat phishing against the identified tools. The approach will use machine-learned algorithms, behavioral analytics, and real-time monitoring to perceive and suspect malicious actions indicative of a phishing process: address deception, false pages of login, and deception of user interactions. In addition, the framework employs novel techniques to detect and prevent unauthorized screenshots, including watermarking, activity tracking, and device identification.