Fescues, which are widely cultivated as grasses and forages around the world, are often naturally infected with the endophyte, Epichloë. This fungus, transmitted through seeds, imparts resistance to drying and herbivorous insects in its host without causing any external damage, thereby contributing to the adaptation of the host to the environment and maintaining a symbiosis. However, some endophytes, such as E. coenophialum synthesize ergovaline or lolitrem B, which accumulate in the plant and impart anti-mammalian properties. For example, when livestock consume excessive amounts of grass containing toxic endophytes, problems associated with neuromuscular abnormalities, such as convulsions, paralysis, high fever, decreased milk production, reproductive disorders, and even death, can occur. Therefore, pre-inoculation with non-toxic endogenous fungi or management with endophyte-free grass is important in preventing damage to livestock and producing high-quality forage. To date, the diagnosis of endophytes has been mainly performed by observation under a microscope following staining, or by performing an immune blot assay using a monoclonal antibody. Recently, the polymerase chain reaction (PCR)-based molecular diagnostic method is gaining importance in the fields of agriculture, livestock, and healthcare given the method’s advantages. These include faster results, with greater accuracy and sensitivity than those obtained using conventional diagnostic methods. For the diagnosis of endophytes, the nested PCR method is the only available option developed; however, it is limited by the fact that the level of toxic alkaloid synthesis cannot be estimated. Therefore, in this study, we aimed to develop a triplex real-time PCR diagnostic method that can determine the presence or absence of endophyte infection using DNA extracted from seeds within 1 h, while simultaneously detecting easD and LtmC genes, which are related to toxic alkaloid synthesis. This new method was then also applied to real field samples.