This dataset accompanies a manuscript submitted to a peer-reviewed scientific journal and extends a previously published study on the mechanism of action of short antimicrobial peptides interacting with lipid vesicles. The present release includes additional computational and experimental data that substantially expand the scope and depth of the original work. The dataset includes results for three short peptides representative of this class of compounds:PEP1: WQWWWWWQW-NH₂ (net charge +0)PEP2: RQWRRWWQR-NH₂ (net charge +4)PEP3: RKIRRKFKK-NH₂ (net charge +7) New coarse-grained molecular dynamics simulations were performed for PEP3, complementing those previously reported for PEP1 and PEP2, and enabling a systematic evaluation of the role of peptide charge, hydrophobicity, and amphipathicity in membrane interaction and disruption. In addition, the dataset incorporates quantum-level analyses, including Quantum Theory of Atoms in Molecules (QTAIM) calculations and molecular electrostatic potential (MEP) maps, providing detailed insight into peptide–lipid interactions. The computational data are complemented by experimental results, including peptide synthesis, antimicrobial activity assays, and the evaluation of key physicochemical parameters known to modulate antimicrobial activity. Together, these data support a mechanistic model in which short antimicrobial peptides follow a mode of action distinct from that of longer sequences, consistent with a three-step process involving membrane adsorption, cooperative accumulation to a critical surface concentration, and the onset of membrane mechanical instabilities leading to bilayer disruption. The dataset is organized into two main directories. The Molecular_Dynamics directory contains all files required to reproduce the simulations, including force-field parameters, initial coordinates, topology files, and input/output files for vesicle and vesicle–peptide systems. The DATA directory includes scripts and corresponding input/output files used for post-processing and analysis with VMD and SuAVE, as well as QTAIM and MEP calculations. This repository is intended to ensure full reproducibility and to facilitate reuse of the data by the community. For further details, please contact Ezequiel Frigini (enfrigini@unsl.edu.ar), Sergio Pantano (spantano@pasteur.edu.uy), Tamás Beke-Somfai (beke-somfai.tamas@ttk.mta.hu) and Ricardo D. Enriz (denriz@unsl.edu.ar).