The goal of this work is to design and implement an agent-based system that facilitates interaction among a group of individuals who are part of an organization, observe types and frequencies of interaction from which we can draw some conclusions about nature of social interactions among individuals and the group, and automate interaction via agents who represent their human counterparts. The resulting system is a test-bed to model a variety of knowledge aware environments but is independent of any domain. Our test bed model is a 50x50 grid, where we randomly place a large number of eggs. Agents collaborate with each other or work independently to capture all the eggs. We show that when agents work in a team they achieve the maximum output as a whole. These findings may also suggest a prevailing model of trust among agents.