Local land-use transitions, e.g. from a diverse to a specialized system, are increasingly dependent on distant drivers. Therefore, the resilience of social-ecological systems (SESs), i.e. their capacity to absorb ongoing change and maintain ecosystem service provision, needs to be evaluated in the light of socioeconomic and environmental interactions across space - so-called telecouplings. In contrast to traditional approaches, Agent-based models (ABMs) can capture dynamic responses to distant drivers across different levels of organization and scales, which offers important insights into resilience. In this study, we reviewed the potentials and needs to investigate the resilience of telecoupled SESs with ABMs. Moreover, we developed a generic model to assess (i) when agricultural systems respond to distant drivers and sending systems emerge, (ii) when this leads to resource use specialization, and (iii) how this subsequently affects system resilience. ABMs have a great potential to deal with the multidimensionality of resilience and complex systems through extensive manipulation in a fully controlled system. To successfully investigate telecoupled systems, models need to systematically consider the relevant dynamics, feedbacks and flows between and within the coupled systems. Multiple patterns observed in real systems at different spatial and temporal scales, and levels of organization will allow us to design, parameterize, and validate such ABMs. Our initial model indicates that while distant demand for agricultural products typically favors profit maximizing agents, cooperation, green subsidies and alternative investments can support smallholders and maintain higher land-use diversity. ABMs provide a way forward in exploring the consequences of telecoupling on the resilience of SESs. Unveiling the importance of different factors in facilitating local responses to distant drivers can help to identify real-world landscapes where future telecouplings may emerge.