Naso-pharyngeal RT-PCR is the gold standard for the diagnosis of COVID-19, but there is a need for rapid and reliable tests. Some validation studies have used frozen aliquots mainly from adults. The aim of this real-life study was to test the performance of a SARS-CoV-2 rapid antigen test (SC2-RAT) in children. Symptomatic patients aged 0 to 17 years were recruited in the emergency department of the University Hospital of Creteil and in primary care pediatric practices from October 10, 2020 for 7 weeks. Each enrolled child had a SARS-CoV-2 RT-PCR test and a SC2-RAT from two distinct nasopharyngeal swabs. Among the 308 patients (mean SD age 4.9 5.3 years), fever was the main symptom (73.4%), with no difference between COVID-19–negative and –positive groups. The prevalence of COVID-19 was 10.7% (95% CI 7.5–14.7). On the whole cohort, the sensitivity and specificity of the SC2-RAT compared to RT-PCR was 87.9% (95% CI 71.8–96.6) and 98.5% (95% CI 96.3–99.6). Considering samples with cycle threshold >25, the sensibility was lower: 63.6% (95% CI 30.8–89.1) and the specificity 99.6% (95% CI 98.0–100.0). The mean delay to obtain an SC2-RAT result was <15 min but was 3.2 h (SD 5.5) for an RT-PCR result. Contact with a COVID-19–positive person was more frequent for COVID-19–positive than –negative patients ( n = 21, 61.6%, vs. n = 64, 24.6%; p < 0.01). In real life, SC2-RAT seems reliable for symptomatic children, allowing to detect contagious children.