This article focuses on the multimodal support of L2 education when learners visit extracurricular learning sites. As part of a design-based research project, a team of students, cooperative partners, and the researcher developed an educational game using Actionbound to enhance on-site cultural and language education. The aim of this paper is to document and analyze the multimodal interplay between the context and the behavior of B1-level L2 German language learners, as guided by digital stimuli during a scavenger hunt. The analysis of the multimodal interplay offers insight into the potential for culturally and linguistically adapted learning support. A corpus of video, photo, and audio recordings is utilized to analyze the multimodal interaction of L2 learners (n = 7; July 2024) as they engaged with the application. The case study of one learning group (n = 3; aged 15, 16, and 17 years) at the secondary school level illustrates how the context of the combined virtual and physical sites is perceived by the learners and to what extent the tasks within this context contribute to linguistic and cultural learning.