The NIME and HCI domains have seen the emergence of several tools for designing expressive gestural interactions with sounds. Each tool has its own specificity, such as high versatility through low-level APIs, or inclusion of non-expert users through high-level components and demonstration-based design workflows. In this short paper, we present a toolkit, called Gestural Sound Toolkit (GST), that we have been using for the past nine years in research projects and teaching. We describe this toolkit and the design process that led to its current status. Our goal is to reflect on the decisions made, and to highlight the factors of these decisions, or their arbitrariness. In doing so, we hope to provide insights for the development of tools for designing gestural interaction with sound.