This study aimed to examine the developmental pattern of diurnal cortisol rhythm during pubertal transition and its prospective association with psychopathological symptoms. A cohort of 1158 children consisting of 608 boys and 550 girls aged 7 to 9 years (mean [standard deviation] age = 8.04 [0.61] years) were recruited in the Anhui Province of China in 2015 (wave 1). A single awakening sample was collected at baseline, and three additional samples were collected at one weekday in wave 2 to wave 4. Four indices of cortisol activity were evaluated and calculated across the day: awakening cortisol level, cortisol awakening response, the area under the curve with respect to ground (AUC), and the diurnal cortisol slope. In each wave, pubertal development was assessed by testicular size in boys and Tanner scales in girls. Psychopathological symptoms were ascertained in waves 2 to 4. Multilevel mixed models revealed no significant pubertal changes in diurnal cortisol activity in girls. In boys, awakening cortisol (β = -0.005, p = .004) and total cortisol output (lnAUC, β = -0.005, p = .040) significantly decreased across pubertal transition. Higher awakening cortisol and total cortisol output (lnAUC) were associated with higher scores on internalizing symptoms in girls (β = 0.82, p < .001; β = 0.62, p = .012) and externalizing symptoms in boys (β = 0.73, p = .001; β = 0.55, p = .019) during the 3-year follow-up. In contrast, no associations were found between cortisol awakening response and diurnal cortisol slope with psychopathological symptom scores in boys or girls. Development of diurnal cortisol activity during pubertal transition occurs in a sex-specific manner. Awakening cortisol level and daily total cortisol output may serve as markers for psychopathology during pubertal transition.