Abstract Introduction Post-stroke aphasia makes it difficult to assess cognitive deficiencies. We thus developed the CASP, which can be administered without using language. Our objective was to compare the feasibility of the CASP, the Mini Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA) in aphasic stroke patients. Material and methods All aphasic patients consecutively admitted to seven French rehabilitation units during a 4-month period after a recent first left hemispheric stroke were assessed with CASP, MMSE and MoCA. We determined the proportion of patients in whom it was impossible to administer at least one item from these 3 scales, and compared their administration times. Results Forty-four patients were included (age 64 ± 15, 26 males). The CASP was impossible to administer in eight of them (18%), compared with 16 for the MMSE (36%, P = 0.05) and 13 for the MoCA (30%, P = 0.21, NS). It was possible to administer the CASP in all of the patients with expressive aphasia, whereas the MMSE and the MoCA could not be administered. Administration times were longer for the CASP (13 ± 4 min) than for the MMSE (8 ± 3 min, P < 10−6 ) and the MoCA (11 ± 5 min, P = 0.23, NS). Conclusion The CASP is more feasible than the MMSE and the MoCA in aphasic stroke patients.