Snake venom consists of toxin proteins with multiple disulfide linkages to generate unique structures and biological functions. Determination of these cysteine connections usually requires the purification of each protein followed by structural analysis. In this study, dimethyl labeling coupled with LC-MS/MS and RADAR algorithm was developed to identify the disulfide bonds in crude snake venom. Without any protein separation, the disulfide linkages of several cytotoxins and PLA2 could be solved, including more than 20 disulfide bonds. The results show that this method is capable of analyzing protein mixture. In addition, the approach was also used to compare native cytotoxin 3 (CTX III) and its scrambled isomer, another category of protein mixture, for unknown disulfide bonds. Two disulfide-linked peptides were observed in the native CTX III, and 10 in its scrambled form, X-CTX III. This is the first study that reports a platform for the global cysteine connection analysis on a protein mixture. The proposed method is simple and automatic, offering an efficient tool for structural and functional studies of venom proteins.