What problem does it solve? Deciding how to allocate OUSD collateral across Ethereum, Base, and HyperEVM Morpho strategies, the Curve AMO, and the vault buffer requires reconciling live on-chain state, transient versus sustainable APYs, and production constraints — a manual process prone to trusting stale or misleading data. ## Core Features & Use Cases - Independent on-chain verification: Reads deployed balances, utilization, and liquidity directly via cast rather than trusting the MCP or the production rebalancer feed. - Sustainable-rate optimization: Values each market at its 90%-utilization rate and applies production gates (0.5% spread floor, $5K/$25K minimums, $50K steps, freeze thresholds) to produce an executable net-moves plan. - Learning loop: Logs every decision, scores past recommendations against realized Grafana history, and accumulates market-sensitivity heuristics in LEARNINGS.md. - Use Case: Ask whether to move funds from Base to HyperEVM and receive a sized, constraint-checked plan with expected $/yr delta, pre-flighted against on-chain liquidity. ## Quick Start Ask the assistant to optimize the OUSD allocation and recommend any rebalancing moves across the Morpho strategies, AMO, and vault buffer.