Biogenic amines containing NH3 are important indicators for conducting full-scale appraisal of food spoilage and disease diagnosis. However, the currently-used detection methods of NH3 have several limitations such as time-consuming high cost, and inability to provide visual real-time monitoring. Therefore, researchers have attempted to explore strategies for quantitative real-time monitoring of NH3 for food spoilage has attracted widespread attentions. Herein, we developed sustainable, fast response, hypersensitized, user-friendly and molecular-level light-emitting biomass-based materials (AFP-FP) containing on-demand nanostructured brain-cells-inspired aggregation-induced-emission (AIE) self-assembles for real-time visual monitoring of seafood spoilage. The 2-hydroxy-5-methyl-isophthalaldehyde-based AIE probe (AFP) was synthesized using a simple “one-step” route. AFP-FP exhibited high selectivity, sensitivity, repeatable and quantitative recognition (y = 7.292×103x + 7.621×104, R = 0.990) of NH3 with a low detection limit (246 ppb) and fast response (<1 s). Furthermore, we integrated AFP-FP into a user-friendly smartphone color recognition app, enabling its practical application in visual, real-time daylight monitoring of food spoilage.