Glycosides make up a biomedically important class of secondary metabolites. Most naturally occurring glycosides were isolated from plants and bacteria; however, the chemical diversity of glycosylated natural products in fungi remains largely unexplored. Herein, we present a paradigm to specifically discover diverse and bioactive glycosylated natural products from fungi by combining tailoring enzyme-guided genome mining with mass spectrometry (MS)-based metabolome analysis. Through in vivo genes deletion and heterologous expression, the first fungal C-glycosyltransferase AuCGT involved in the biosynthesis of stromemycin was identified from Aspergillus ustus. Subsequent homology-based genome mining for fungal glycosyltransferases by using AuCGT as a probe revealed a variety of biosynthetic gene clusters (BGCs) containing its homologues in diverse fungi, of which the glycoside-producing capability was corroborated by high-performance liquid chromatography-mass spectrometry (HPLC-MS) analysis. Consequently, 28 fungal aromatic polyketide C/O-glycosides, including 20 new compounds, were efficiently discovered and isolated from the three selected fungi. Moreover, several novel fungal C/O-glycosyltransferases, especially three novel α-pyrone C-glycosyltransferases, were functionally characterized and verified in the biosynthesis of these glycosides. In addition, a proof of principle for combinatorial biosynthesis was applied to design the production of unnatural glycosides in Aspergillus nidulans. Notably, the newly discovered glycosides exhibited significant antiviral, antibacterial, and antidiabetic activities. Our work demonstrates the promise of tailoring enzyme-guided genome-mining approach for the targeted discovery of fungal glycosides and promotes the exploration of a broader chemical space for natural products with a target structural motif in microbial genomes.