Background <p>Physical fitness in children and adolescents is an important outcome of school health promotion. Previous reviews have often combined physical education, recess, after-school programmes, and whole-school strategies, making it difficult to determine whether embedded fitness-oriented content improves multidimensional physical fitness. This study evaluated the effects of physical education-embedded fitness-oriented exercise interventions on physical fitness in general populations of children and adolescents.</p> Methods <p>The initial systematic search was completed on March 11, 2026, in PubMed, Web of Science Core Collection, Embase, Cochrane Library/CENTRAL, CINAHL, SPORTDiscus, Scopus, CNKI, Wanfang, and VIP. Randomized controlled trials (RCTs) and cluster-randomized controlled trials (cRCTs) conducted among children and adolescents in general schools were included. Interventions were fitness-oriented exercise content embedded within regular physical education lessons, warm-ups, or main lesson components; comparators were regular physical education, usual school physical activity, usual warm-up, or wait-list control. Outcomes were lower-limb explosive power, muscular endurance, speed, agility, and flexibility. Hedges g was pooled using random-effects models. Effective sample size correction was applied to cRCTs that did not adequately account for clustering. RoB 2 and GRADE were used to assess risk of bias and certainty of evidence.</p> Results <p>Formal search updates and targeted supplementary searches identified 7,875 records. After deduplication and removal of overlaps between search sources, 4,724 records entered title and abstract screening. Finally, 54 reports representing 51 independent studies were included in the systematic review, and 49 studies entered the primary arm-level Hedges g meta-analysis. The primary analysis showed favourable average effects for lower-limb explosive power (Hedges g = 0.260, 95% CI 0.081 to 0.438), muscular endurance (g = 0.396, 95% CI 0.237 to 0.554), agility (g = 0.494, 95% CI 0.151 to 0.837), and flexibility (g = 0.505, 95% CI 0.106 to 0.905), with 95% CIs not crossing the line of no effect. Speed also showed a favourable direction, but the 95% CI crossed the line of no effect (g = 0.198, 95% CI − 0.127 to 0.522). The certainty of evidence was low for muscular endurance and very low for the other four outcomes.</p> Conclusions <p>Low- to very-low-certainty evidence suggests that physical education–embedded fitness-oriented exercise interventions may have small favourable effects on some physical fitness outcomes in children and adolescents. Given the risk of bias, heterogeneity, imprecision, and small-study effects or funnel plot asymmetry signals in some outcomes, the current results should not be interpreted as definitive evidence of effectiveness. These interventions may represent a low-additional-time pathway for improving the quality of fitness stimuli within existing physical education lessons, but current evidence remains insufficient to support strong practice recommendations.</p> Trial registration <p>Systematic review registration: PROSPERO CRD420261356863.</p>