The development of tools and methods supporting the identification of Industrial Symbiosis opportunities is of utmost importance to unlock its full potential. Knowledge repositories have proven to be powerful tools in this sense, but often fail mainly due to poor contextualization of information and lack of general applicability (out of the boundaries of specific areas or projects). In this work, a novel approach to the design of knowledge repositories for Industrial Symbiosis is presented, based on the inclusion and categorization of tacit knowledge as well as on the combination of mimicking and input-output matching approaches. The results of a first usability test of the proposed tool are also illustrated.