That corals skeletons are built of aragonite crystals with taxonomy-linked ultrastructure has been well understood since the 19th century. Yet, the way by which corals control this crystallization process remains an unsolved question. Here, I outline a new conceptual model of coral biomineralisation that endeavours to relate known skeletal features with homeostatic functions beyond traditional growth (structural) determinants. In particular, I propose that the dominant physiological driver of skeletal extension is night-time hypoxia, which is exacerbated by the respiratory oxygen demands of the coral's algal symbionts (= zooxanthellae). The model thus provides a new narrative to explain the high growth rate of symbiotic corals, by equating skeletal deposition with the "work-rate" of the coral host needed to maintain a stable and beneficial symbiosis. In this way, coral skeletons are interpreted as a continuous (long-run) recording unit of the stability and functioning of the coral–algae endosymbiosis. After providing supportive evidence for the model across multiple scales of observation, I use coral core data from the Great Barrier Reef (Australia) to highlight the disturbed nature of the symbiosis in recent decades, but suggest that its onset is consistent with a trajectory that has been followed since at least the start of the 1900s. In concluding, I outline how the proposed capacity of cnidarians (which includes modern reef corals) to overcome the metabolic limitation of hypoxia via skeletogenesis also provides a new hypothesis to explain the sudden appearance in the fossil record of calcified skeletons at the Precambrian–Cambrian transition – and the ensuing rapid appearance of most major animal phyla