Autonomous agents interacting in an open world can be considered to be primarily driven by self interests. Previous work in this area has prescribed a strategy of reciprocal behavior for promoting and sustaining cooperation among self-interested agents. In an extension to that work the importance of using other’s "opinion" by the reciprocative agents to effectively curb the exploitative tendencies of selfish agents has also been evaluated. However, one of the restrictions so far has been the metric of "cost" evaluation. So far we have considered only the time of completion of a task to be the measure of the task cost. In this paper, we expand the cost metric by including both time of delivery and quality of work. This allows us to model more realistic domains. In contrast to our previous work, we also allow for agents with varying expertise for different job types. This necessitates the incorporation of the novel aspect of learning about other’s capabilities within the reciprocity framework. In this paper, we evaluate the hypotheses that self-interested agents with complementary expertise can learn to recognize cooperation possibilities and develop stable, mutually beneficial coalitions that is resistant to exploitation by malevolent agents.