MINFLUX microscopy allows characterization of molecular organization and dynamics with single nanometer spatial resolution and sub-hundred microseconds temporal resolution. However, acquisition times often span minutes to hours as a single fluorophore is measured at a time. Studying live cellular processes therefore requires careful consideration of where and when to apply it, hence manual control limits its potential applications. To overcome acquisition speed, initiation, and data throughput limitations, we present event-triggered MINFLUX: a smart microscopy method using confocal monitoring with real-time image analysis, and applying MINFLUX exactly where and when deemed necessary. The method is controlled through a custom-written open-source Python framework automatically controlling a commercial MINFLUX microscope. We investigate molecular membrane dynamics and organization in 2D and 3D during cellular events: lipid dynamics at caveolae; membrane topography during dynamin-mediated endocytosis; and membrane fluidity and topography during HIV-1 budding site formation. Rapid event detection and minimal regions of interest provides data that would be unfeasible or impossible to acquire through manual microscope control.