COVID-19 pandemic has pointed out the strong need to continue promoting the development of innovative approaches for the rapid, safe handling and accurate molecular diagnosis of SARS-CoV-2 infection. In this work, we have developed CoVradar, a novel colourimetric molecular detection system that integrates nucleic acid analysis by dynamic chemistry labelling (DCL) technology with the Spin-Tube devices allowing reverse dot-blot hybridization in an array format to detect SARS-CoV-2 RNA within saliva specimens. The assay is designed so that ten regions per RNA copy act as templates to amplify the signal generated through a DCL reaction. The sequences are captured by abasic peptide nucleic acids (DGL-Probes) immobilised onto nylon membranes with a specific spot pattern. Biotinylated SMART-Bases are used for the labelling of duplexes formed upon the capture of complementary ssRNA fragment sequences that act as templates. Signals are then generated by the recognition of the biotin via streptavidin alkaline phosphatase and incubation with a chromogenic substrate which generates a blue precipitate. CoVradar’s results are automatically interpreted by CoVreader, a dedicated smartphone-based Image Processing System (IPS) capable of imaging and analysing the spot pattern developed to record and share results by mobile phone. The combination of CoVradar and CoVreader delivers a unique molecular assay capable of detecting SARS-CoV-2 viral RNA with single base specificity. This represents an innovative solution providing benefits in terms of time, cost, and simplicity, all of which are crucial for the diagnosis of COVID-19 but also in time, a repertoire of assays for other infectious diseases.