Rabies is a zoonotic and fatal encephalitis caused by members of the Lyssavirus genus. Among them, the most relevant species is Lyssavirus rabies, which is estimated to cause 60,000 human and most mammal rabies deaths annually worldwide. Nevertheless, all lyssaviruses can invariably cause rabies, and therefore their impact on animal and public health should not be neglected. For accurate and reliable surveillance, diagnosis should rely on broad-spectrum tests able to detect all known lyssaviruses, including the most divergent ones. In the present study, we evaluated four different pan-lyssavirus protocols widely used at an international level, including two real-time RT-PCR assays (namely LN34 and JW12/N165-146), a hemi-nested RT-PCR and a one-step RT-PCR. Additionally, an improved version of the LN34 assay ((n) LN34) was developed to increase primer–template complementarity with respect to all lyssavirus species. All protocols were evaluated in silico, and their performance was compared in vitro employing 18 lyssavirus RNAs (encompassing 15 species). The (n) LN34 assay showed enhanced sensitivity in detecting most lyssavirus species, with limits of detection ranging from 10 to 100 RNA copies/µL depending on the strain, while retaining high sensitivity against Lyssavirus rabies. The development of this protocol represents a step forward towards improved surveillance of the entire Lyssavirus genus.