Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) nucleic acid detection is the gold standard for the laboratory diagnosis of coronavirus disease 2019 (COVID‐19). However, this method has high requirements for practitioners' skills and testing sites, so it is not easy to popularize and promote the application in places other than large hospitals. In addition, the detection flux of SARS‐CoV‐2 nucleic acid is small, and the whole detection process takes much time, which cannot meet the actual needs of rapid screening in large quantities. The WHO conditionally approved a batch of SARS‐CoV‐2 antigen reagents for clinical application to alleviate this contradiction. SARS‐CoV‐2 antigen detection offers a trade‐off among clinical performance, speed and accessibility. With the gradual increase in clinical application, the accumulated clinical data show that the sensitivity and specificity of the SARS‐CoV‐2 antigen assay are over 80% and 97%, respectively, which can basically meet the requirements of the WHO. However, the sensitivity of the SARS‐CoV‐2 Antigen Assay among asymptomatic people in low prevalence areas of COVID‐19 cannot meet the standard, leading to a large number of missed diagnoses. In addition, the detection ability of SARS‐CoV‐2 antigen reagent for different SARS‐CoV‐2 mutant strains differs greatly, especially for those escaping the COVID‐19 vaccines. In terms of results interpretation, it is highly reliable to exclude SARS‐CoV‐2 infection based on the high negative predictive value of the SARS‐CoV‐2 antigen assay. However, in the low prevalence environment, the probability of false positives of the SARS‐CoV‐2 antigen assay is high, so the positive results need to be confirmed by the SARS‐CoV‐2 nucleic acid reagent. The SARS‐CoV‐2 antigen assay is only a supplement to SARS‐CoV‐2 nucleic acid detection and can never completely replace it. To date, SARS‐CoV‐2 nucleic acid detection continues to be the standard laboratory method for COVID‐19 diagnosis.