Highly pathogenic avian influenza H5N1 virus causes heavy losses in poultry farms worldwide. Molecular diagnostic techniques like RT-PCR and real-time RT-PCR are considered the gold standard for identification of H5 influenza viruses in clinical samples. These techniques are hampered by the need of well-equipped laboratories, large space requirement, and relatively long time-to-result. Recombinase polymerase amplification (RPA) assay represents an excellent alternative to PCR since it is more simple, rapid, economic, and portable. Reverse transcription RPA (RT-RPA) assay was recently developed for sensitive and specific detection of H5N1 virus in 6-10?min. To ensure the accuracy of the developed assay, two approaches for using a positive control were evaluated in this study. These approaches included: 1) all-in-one (internal positive control; IPC), 2) two-tubes-per-one-sample (external positive control; EPC). Sigma virus (SIGV) RNA and turkey mitochondrial DNA were tested as positive controls in both approaches. For all-in-one approach, both targets (H5 and IPC) were strongly inhibited. In contrast, very good amplification signals were obtained for the two types of EPC with no effect on the analytical sensitivity and specificity of H5 RT-RPA assay in two-tubes-per-one-sample approach. The performance of EPC-based H5 RT-RPA was further validated using 13 tracheal swabs. The results were compared to real-time RT-PCR and proved superior specificity in detecting H5N1 but not H5N8 viruses. Inclusion of EPC did not affect the aptitude of both assays in terms of sensitivity, specificity and reproducibility. In conclusion, the two-tubes-per-one-sample approach was more reliable to control the false negative results in H5 RT-RPA assay.