Objective: On 4 March 2022, the first community-acquired case of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was reported in Vanuatu, with community transmission occurring subsequently. It was expected that the number of notified SARS-CoV-2 cases would be an underestimate of the true infection rate of this outbreak; however, the magnitude of underreporting was unknown. The purpose of this study was to provide a population-based estimate of SARS-CoV-2 infection shortly after the first reports of community transmission, to understand the level of underdetection and undernotification in Vanuatu and thus to inform ongoing prevention and response activities. Methods: We conducted a cross-sectional SARS-CoV-2 prevalence study in two geographical administrative areas in Port Vila, Vanuatu in April 2022. All residents in selected areas were eligible. Trained teams conducted demographic and behavioural interviews and collected nasal specimens. Specimens were tested by polymerase chain reaction. The primary outcomes were the rates of SARS-CoV-2 attack (point prevalence) and cumulative attack, underdetection, notification and household secondary attack. Results: A total of 252 people from 84 households participated. Among 175 people who had a sample collected, 91 were SARS-CoV-2-positive (attack rate 52.0%). Most cases had not been detected before the study (underdetection rate 91.5%). More than half of previously detected cases were notified (notification rate 65.2%). Discussion: Within the first few weeks of community transmission, more than half of participants in the selected areas had evidence of SARS-CoV-2 infection; however, most infections had been undetected. This study provides important information about the rapid spread of novel infectious diseases in Vanuatu.