Clusters of firms are considered by many companies as an effective tool in the system of management of social economic development. The important element in management of cluster development is identification, diagnostics and description of clusters developed in economic environment. Now procedures of identification are labor-consuming and inexact. It happens because of the fact that clusters have indistinct and mobile borders and configurations. Presumably, the efficiency of identification procedures can be increased by using the categorial model "Final information flow". The use of it lets achieve a complex of quality characteristics of clusters of firms reflecting logic of their natural process of evolution, and on their bases to carry out procedures mentioned above. The basic elements of the categorial model "Final information flow" (FIF) are information criteria (IC), logic level (LLv), logic limit (LLm), and transformation (T). In clusters of firms during the process of their natural evolution new quality characteristics appear which are reflected in IC and accordingly form new LLv. Actually it marks the next step in clusters evolution. The following logic levels were revealed for clusters of firms: LL1 - production, LL2 - support, LL3 -maintenance, LL4 - consumption, LL5 - research, LL6 - export, LL7 - training. LLm reflects the limited amount of quality characteristics of a cluster of firms which it has at the existing LLv. At each LLv a cluster can have only characteristics obtained before. In the limits of each LLv many various cluster forms can appear, however, they are limited by the basic set of cluster's quality characteristics, which is reflected by a corresponding LLm. The possibilities of combining quality characteristics in the limits of LLm are reflected by T. The method offered and the results received on its bases on subject domain of clusters of firms allow carrying out a more productive identification of clusters of firms, their quality description and initial diagnostics. The given thesis is confirmed by the example of identification of Omsk agricultural cluster. There were examined the logic levels developed by the present moment, and that allowed unequivocally to define it by carrying out firms which are the carriers of quality characteristics in the limits of each LLv. Besides, it has obtained the quality description of the given cluster and has carried out initial diagnostics, which allowed qualifying it as a developed industrial cluster.