Traditionally, privacy breaches have been understood as the malfunctioning of a given system. However, in online social networks, privacy breaches are not necessarily a malfunctioning of a system but a byproduct of its workings. In online social networks, users are allowed to create and share content about themselves and others. When multiple entities start distributing content without a control, information can reach unintended individuals and inference can reveal more information about the user. To overcome the privacy violations an approach based on agent-based representation of a social network, where the agents manage users privacy requirements by creating commitments with the system is proposed. The proposed detection algorithm performs reasoning using the description logic and commitments on a varying depths of social networks.