Skyfall AI has released MORPHEUS, a persistent enterprise simulation benchmark for continual reinforcement learning (CRL) that does not reset after episodes. MORPHEUS is grounded in the Big World Hypothesis (Javed & Sutton, 2024) and requires three properties: persistence, non-stationarity, and operational complexity. Each environment is a self-contained TypeScript world plugin exporting Operational Descriptors (ODs). Non-stationarity comes from a failure injection engine drawing from eleven failure types at four preset rates: light (5%), realistic (8%), moderate (15%), and aggressive (30%), and an asynchronous configuration shift controller. Reward comes from three operational verifiers: failure event signals, financial ledger status, and resource throughput, with default weights of w_f = 0.5 and w_l = w_p = 0.25. Under upper-bound assumptions, the bound per configuration equals 0.50. For policy initialisation, a frontier model (Gemini 3.1 pro) collects trajectories using the ReAct framework, which then fine-tune Qwen3-14B via supervised fine-tuning (SFT), and every RL run starts from this shared SFT checkpoint.