Cloud federation allows cloud providers to dynamically use resources of other federated providers in order to fulfill the requirements of customer requests. This concept enables the federated cloud providers to use external resources for increasing their profit as they do not have to reject customers in case their own resources are occupied. However, (i) comparing the offers of the federated providers in order to decide which provider to use as well as (ii) adapting the installation scripts of the components to be deployed for the different providers is complex, error-prone, and time consuming. In this paper, we present an approach that enables customers to describe their desired application deployments in the form of a topology model that is independent of any concrete provider. We show how this model can be automatically adapted by a provider participating in a cloud federation to deploy components on different other participants. To ensure the practical feasibility of the approach, we employ the TOSCA standard for describing these models and present a technical system architecture based on existing technologies.