Background: Anxiety and depression are among the most prevalent emotional symptoms in adulthood, often leading to impaired functioning and long-term mental health challenges. While prior research has identified univariate associations of experiential avoidance or self-esteem with these symptoms, less attention has been paid to how these factors jointly shape individual risk profiles. Acceptance and Commitment Therapy (ACT) conceptualizes experiential avoidance and psychological inflexibility as key processes that increase vulnerability, while self-concept theory highlights self-esteem as a protective resource. To capture the heterogeneity within individual psychological profiles, a person-centered approach is warranted. Methods: A community sample of 134 adults residing in the Seoul metropolitan area completed four validated instruments: the Acceptance and Action Questionnaire-II (AAQ-II), the Rosenberg Self-Esteem Scale (RSES), the Generalized Anxiety Disorder-7 (GAD-7), and the Patient Health Questionnaire-9 (PHQ-9). Standardized AAQ-II and RSES scores were subjected to K-means clustering to identify subgroups. The optimal number of clusters was determined based on internal validation indices including the silhouette coefficient, Calinski–Harabasz index, and Davies–Bouldin index. Group differences in anxiety and depression were examined using ANOVA with Tukey HSD post-hoc tests, and supplementary regressions were conducted to assess the incremental explanatory value of cluster membership. Results: Three clusters were identified: Resilient (n = 69), Moderate (n = 34), and Vulnerable (n = 31). Internal indices suggested modest separation (silhouette = .36, CH = 85.36, DB = .95). Anxiety and depression levels differed significantly across groups (GAD-7: F(2, 131) = 13.69, p < .001, η² = .173; PHQ-9: F(2, 131) = 18.44, p < .001, η² = .220). Post-hoc comparisons showed that both Vulnerable and Moderate clusters reported higher symptom scores than the Resilient cluster, while the difference between Vulnerable and Moderate clusters was not significant. Supplementary regressions confirmed positive and significant coefficients for both non-resilient clusters, with model R² values around .165–.167. Conclusion: Findings suggest that combined profiles of experiential avoidance and self-esteem meaningfully differentiate adult emotional risk patterns. Although the classification has limited separation and is constrained by non-probability sampling, cross-sectional design, and self-report data, the typology provides a useful exploratory basis for screening and tailored interventions. Future studies should replicate with larger, stratified samples, longitudinal and interventional studies, and external validation through bootstrap-based consensus clustering.