Industry cluster as a new form of industrial organization has become the dominant model of regional economic development. From foreign development experiences of regional economic, KIBS have performed important roles in the industry clusters. And the number of agencies related to KIBS is constantly increased. In the paper, the method of social network analysis has uses to research the embedding of KIBS in the industrial clusters. The analysis has been gone on from the "relationship" and "structural" aspects. Due to the purpose of this study, network centricity、network architecture degree and network density have been used to describe the cluster network structure. During the research, the cluster network structure as a variable has taken into account. Its action to the growth of the cluster firms has been achieved through intermediate variable. With a view to the meaning of the growth of the cluster firms, the variable of knowledge ability has been used to analyze the cluster network structure. So the variable of knowledge ability becomes an intermediate variable. Knowledge ability is the ability that a firm can continuously learn knowledge or information from the outside in order to achieve coordinated development with the external environment. The cluster knowledge abilities have included knowledge-absorbing ability and knowledge-overflow ability from the view of industrial clusters. Through research we have proposed three levels of 11 assumptions. The knowledge-absorbing ability and knowledge-overflow ability are positively correlated to the growth of cluster firms. In the follow-up study, the author would refine the variables, design questionnaire and statistical verification of these assumptions. The analysis would fully reveal the relationship among the three variables of " Embedding of KIBS——Cluster Network Structure——the Growth of Cluster Firms" according to various statistical results.