Although RNA-seq data are becoming more widely available for biomedical research, most datasets for short non-coding RNAs (sncRNAs) primarily focus on microRNA analysis using standard RNA-seq, which captures only sncRNAs with 5'-phosphate (5'-P) and 3'-hydroxyl (3'-OH) ends. Standard RNA-seq fails to sequence sncRNAs with different terminal phosphate states, including tRNA halves, the most abundant class of tRNA-derived sncRNAs that play diverse roles in various biological processes. tRNA halves are produced through the endoribonucleolytic cleavage of mature tRNA anticodon loops. The responsible endoribonucleases, such as Angiogenin, commonly leave a 2',3'-cyclic phosphate (cP) at the 3'-end of 5'-tRNA halves and forms a 5'-OH end of 3'-tRNA halves, making them incompatible with standard RNA-seq. We developed a method named "cP-RNA-seq" that selectively amplifies and sequences tRNA halves and other cP-containing sncRNAs. Here we describe a detailed and recently updated cP-RNA-seq protocol. In this method, the 3'-end of all sncRNAs, except those containing a cP, are cleaved through periodate treatment after phosphatase treatment. Consequently, adaptor ligation and cDNA amplification steps are exclusively applied to cP-containing sncRNAs. Our cP-RNA-seq only requires commercially available reagents and is broadly applicable for the global identification of tRNA halves and other cP-containing sncRNA repertoires in various transcriptomes.