Developing non-invasive, practical methods to identify exercise metabolic thresholds remains a key focus in exercise physiology. This study examined the variation patterns of average skin temperature (aTsk) measured by infrared thermography during a ramp incremental exercise test (RIET) and determined skin temperature thresholds (TSKTs) to assess their agreement with ventilatory thresholds (VTs). Twenty-three active college students performed a RIET. Physiological parameters including aTsk, oxygen uptake (V˙O2), heart rate (HR), and power output (PO), were recorded. Paired-sample t-tests, Pearson correlation (r), Intraclass Correlation Coefficient (ICC) and Bland-Altman were used to assess the agreement between TSKTs and VTs. The TSKTs comprised the first (TSKT1) and second (TSKT2) skin temperature thresholds. 1) aTsk remained stable initially, then declined (TSKT1), and finally slowed or rebounded (TSKT2). 2) No significant differences were observed between VTs and TSKTs in V˙O2, HR, or PO (P > 0.05). 3) At the levels of absolute V˙O2 (V˙O2ab), relative V˙O2 (V˙O2re), HR and PO, TSKT1 exhibited moderate agreement with VT1, with ICC (0.587, 0.432, 0.430 and 0.525) and r > 0.40 (P < 0.05). TSKT2 showed moderate to high agreement with VT2 for V˙O2ab (ICC: 0.774, r: 0.778) (P < 0.01), V˙O2 re (ICC: 0.433, r: 0.449) (P < 0.05), and PO (ICC: 0.770, r: 0.784) (P < 0.01), while no significant agreement and correlation was found for HR (P > 0.05). TSKTs identified during RIET exhibited moderate to high reliability in estimating VTs and may serve as a potential tool for VTs assessment.