Aim: To investigate the interrater and test–retest reliability of the Assisting Hand Assessment in adolescents (Ad-AHA) with cerebral palsy (CP) and to evaluate the alternate-form reliability of different test activities. Method: Participants were 112 adolescents with unilateral CP (60 males, 52 females; mean age 14y 5mo [standard deviation SD 2y 8mo], Manual Ability Classification System levels I–III). Reliability was evaluated using intraclass correlation coefficients (ICC), smallest detectable change (SDC), and Bland–Altman plots. Results: ICCs for interrater (n=38) and test–retest reliability (n=31) were excellent: 0.97 (95% CI 0.94–0.98) and 0.99 (95% CI 0.98–0.99) respectively. The alternate-form reliability of different test activities was excellent for children (age 10–12y, n=30) performing the School-Kids AHA and Ad-AHA Board Game 0.99 (95% CI 0.98–0.99) and for adolescents (age 13–18y) performing the Ad-AHA Board Game compared to the Ad-AHA Present (n=28) 0.99 (95% CI 0.95–0.98), or the Ad-AHA Sandwich (n=29) 0.99 (95% CI 0.98–0.99) tasks. SDC for test–retest was 4.5 AHA-units. Interpretation: Ad-AHA scores are consistent across different raters and occasions. The good alternate-form reliability indicates that the different test activities can be used interchangeably in adolescents with unilateral CP. Differences greater than or equal to 5 AHA-units can be considered a change beyond measurement error. The use of logit based AHA-units makes change comparable for persons at different ability levels.