Nope, you're not imagining things. The article actually quite wrong here: kilonovae are so named because they yield ~1000 times the peak luminosity of classical

Nope, you're not imagining things. The article actually quite wrong here: kilonovae are so named because they yield ~1000 times the peak luminosity of classical novae (~1e8 Lsun vs ~1e5 Lsun), not supernovae. For that matter, supernovae are themselves also named that way because their peak luminosities far exceed that of classical novae (with the most luminous recorded being >1e11 Lsun). I am not an expert on magnetar mergers in particular but according to the MNRAS paper that is cited, the transient in question, AT 2023fhn, gave a peak luminosity of about 4e8 Lsun (under certain assumptions about wavelength dependence, distance, etc) --- nowhere near supernova territory. Notably, the paper doesn't actually make the same claim about magnetars that the HN-posted article does.