The goal of this study was to translate and adapt the original 9-item of the Contextual Assessment of Social Skills (CASS) to a Dutch version and assess its psychometric qualities. Autistic adolescents aged 12 to 18 years (n = 99) took part in a randomized controlled trial. In this study, pre-intervention data were utilized. The original CASS was adapted to ensure cultural relevance and the content validity was assessed. Data was used to assess reliability and structural validity, using confirmatory factor analysis. 4-item were added to the CASS during the adaptation to better align with the objectives of the experimental intervention. The original 9-item had inter-item correlations between 0.01 and 0.70. The Cronbach's alpha for the original 4-item total score was moderate ([alpha] = 0.69), while for a 7-item total score, it was high ([alpha] = 0.86). This 7-item total score had a sufficient model fit (Comparative Fit Index = 0.90). This total score had a significant correlation with the Assertion subscale of the Social Skills Improvement System-Adolescent (SSIS-A) (r = 0.26, p < 0.01), and the Social Responsiveness Scale-2 (SRS-2) total score (r = - 0.21, p = 0.04) indicating sufficient convergent validity. The CASS total score was not correlated with the Repetitive and Restricted Behavior scale of the SRS-2 (r = - 0.08, p = 0.43), indicating sufficient divergent validity. The Dutch CASS can be considered a conceptually sound and reliable observational instrument for assessing social conversational skills in Dutch autistic youth. Further evaluation of its feasibility when implemented in practice, outside of clinical research, is needed.