In-community non-invasive identification of asthma-specific volatile organic compounds (VOCs) in exhaled breath presents opportunities to characterize phenotypes, and monitor disease state and therapies. The feasibility of breath sampling with children and the preliminary identification of childhood asthma markers were studied. End-tidal exhaled breath was sampled (2.5 dm³) from 11 children with asthma and 12 healthy children with an adaptive breath sampler. VOCs were collected onto a Tenax^®/Carbotrap hydrophobic adsorbent trap, and analyzed by GC–MS. Classification was by retention-index and mass spectra in a ‘breath matrix‘ followed by multivariate analysis. A panel of eight candidate markers (1-(methylsulfanyl)propane, ethylbenzene, 1,4-dichlorobenzene, 4-isopropenyl-1-methylcyclohexene, 2-octenal, octadecyne, 1-isopropyl-3-methylbenzene and 1,7-dimethylnaphtalene) were found to differentiate between the asthmatic and healthy children in the test cohort with complete separation by 2D principal components analysis (2D PCA). Furthermore, the breath sampling protocol was found to be acceptable to children and young people. This method was found to be acceptable for children, and healthy and asthmatic individuals were distinguished on the basis of eight VOCs at elevated levels in the breath of asthmatic children.