Abstract: Over the past decade, the way people communicate, share content and eventually conduct business has been disrupted by social networking platforms. By bringing together pervasive social computing with social networking under a common modelling methodology and common social rules based on human activities in accordance with the Activity Theory, our work aims at contributing to the standardisation of the Social Web in a way that is independent from proprietary solutions, interoperable and semantically rich for applications beyond social networking. The concept of a Social Agent is introduced as a set of micro-services to handle digital traces of human activities for further analysis, respecting both users’ privacy and the business models of companies that build personalised services or applications in the web. In this paper, the landscape of social networking, social pervasive computing and social agency is presented in detail, the research questions driving our work are explained and the fundamentals of a Social Agent along with the pervasive social network it may build are set. A quantitative evaluation based on the Technology Acceptance Model methodology is implemented on a sample of 200 interviewees, evaluating the concepts of the Social Agent, while its technical feasibility is proven and presented through a prototype.