This study aims to analyze serum protein changes in patients with myasthenia gravis (MG) via proteomics to identify and validate potential biomarkers. Methods: Serum samples were collected from 10 MG patients and 10 healthy controls. The data-independent acquisition (DIA) quantitative proteomics technique was used to measure serum protein levels and identify differentially expressed proteins. Functional enrichment analyses, including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene set enrichment analysis (GSEA), were performed to determine the enrichment trends of the differentially expressed proteins. On the basis of the p values from the enrichment analysis, hierarchical clustering was used to identify significantly enriched pathways in each group. The STRING database was used to construct protein–protein interaction networks of the differentially expressed proteins, identifying the top 50 most closely interacting protein subnetworks and their associated pathways. The target proteins were then validated via mass spectrometry-based targeted proteomics (PRM). Receiver operating characteristic (ROC) curves were plotted for visual analysis. Additionally, serum samples from 20 MG patients and 20 healthy controls(HCs) were collected to validate the proteomic results via enzyme-linked immunosorbent assay (ELISA). Results: A total of 220 differentially expressed proteins were identified between the MG and healthy control groups, with 49 upregulated proteins and 171 downregulated proteins. The differentially expressed proteins were primarily involved in the complement coagulation cascade, tight junctions, actin cytoskeleton regulation, and the Rap1 signaling pathway. PRM validation confirmed the involvement of five proteins—protein S (PROS1), protein C (PROC), C4b-binding protein alpha chain (C4BPA), profilin-1 (PFN1), and talin-1 (TLN1)—in the coagulation-complement-inflammation-cytoskeleton damage cascade. ROC curve analysis revealed that the area under the curve (AUC) for PROS1, PROC, C4BPA, PFN1, and TLN1 was greater than 0.8, indicating their strong diagnostic potential. ELISAs revealed significantly elevated levels of PROS1, PROC, and C4BPA in the MG group compared with the control group (p < 0.01) and significantly decreased levels of PFN1 and TLN1 (p < 0.05, p < 0.01). Conclusion: PROS1, PROC, C4BPA, PFN1, and TLN1 may serve as novel biomarkers for the early screening of Myasthenia Gravis.