This paper presents a high-frequency structural VAR framework for identifying oil price shocks and examining their uncertainty transmission in the U.S. macroeconomy and financial markets. Leveraging the stylized features of financial data - specifically, volatility clustering effectively captured by a GARCH model - this approach achieves global identification of shocks while allowing for volatility spillovers across them. Findings reveal that increased variance in aggregate demand shocks increases the oil-equity price covariance, while precautionary demand shocks, triggering heightened investor risk aversion, significantly diminish this covariance. A real-time forecast error variance decomposition further highlights that oil supply uncertainty was the primary source of oil price forecast uncertainty from late March to early May 2020, yet it contributed minimally during the 2022 Russian invasion of Ukraine.