Autonomic computing has become an important paradigm for dealing with large scale network management. However, changes operated by administrators and self-governed entities may generate vulnerable configurations increasing the exposure to security attacks. In this paper, we propose a novel approach for supporting collaborative treatments in order to remediate known security vulnerabilities in autonomic networks and systems. We put forward a mathematical formulation of vulnerability treatments as well as an XCCDF-based language for specifying them in a machine-readable manner. We describe a collaborative framework for performing these treatments taking advantage of optimized algorithms, and evaluate its performance in order to show the feasibility of our solution.