Background <p>Skin cutaneous melanoma (SKCM) represents the most aggressive type of skin cancer. Accumulating evidence underscores the critical involvement of chronic psychological stress and depressive states in cancer development. Genes linked to psychological stress pathways are known to regulate tumor dynamics and may serve as prognostic indicators. However, their precise implications for SKCM patient outcomes and the efficacy of immunotherapeutic interventions remain to be fully elucidated.</p> Methods <p>Prognostically significant psychological stress-related genes (PSRGs) were first screened via univariate Cox regression in The Cancer Genome Atlas (TCGA) cohort. An optimal gene signature was then developed using least absolute shrinkage and selection operator regression, followed by multivariate Cox analysis. The prognostic accuracy of the signature was evaluated through Kaplan–Meier survival curves, area under the curve (AUC) values, and calibration plots. To uncover associated biological processes, pathway and immune infiltration analyses were performed comparing groups stratified by signature risk scores.</p> Results <p>From an initial panel of 374 PSRGs, 64 were significantly correlated with overall survival (OS). A 12-gene prognostic signature was ultimately constructed. The model demonstrated AUCs of 0.766 and 0.644 for 5-year OS prediction in the TCGA training set and the GSE65904 validation set, respectively. A lower signature score was associated with improved survival outcomes, an immunologically active tumor microenvironment enriched for CD8⁺ T cell infiltration, and enhanced likelihood of response to immunotherapy.</p> Conclusion <p>This study establishes a 12-gene signature based on PSRGs, which reliably forecasts clinical prognosis in SKCM. The signature captures features of an immune-activated microenvironment, indicating its dual utility for refining risk stratification and predicting benefit from immunotherapy.</p>