Checked out Spark's buyback proposal: The core design is to establish a DAO Proxy treasury through protocol revenue, which includes risk reserves, product loss reserves, and operational reserves. The protocol calculates a target value. The buyback is triggered when accumulated revenue exceeds this target value. Each month, 10% of the excess is used to buy back SPK. If the revenue exceeds 200% of the target value, the entire excess is used for buybacks. The calculation of the DAO Proxy target value is divided into two parts. Risk reserves and product loss reserves are categorized as capital reserve requirements, while operational reserves are another part. The final target value is determined by taking the maximum of the two rather than summing them up. Risk reserves are basic parameters set by Sky, serving as fundamental protection for the protocol. Product loss reserves act as a risk buffer for products like Spark Savings and SparkLend, and can be used to cover potential losses or bad debts when necessary. Currently, Spark SLL's estimated annual revenue is $24 million, based on the current market size and interest rate environment. The regular buyback uses 10% of the funds exceeding the target value, which is relatively conservative. However, if the accumulation exceeds twice the target value, the buyback is intensified, and the enhanced buyback scale is quite significant. So the overall design prioritizes meeting the protocol's safety reserves, then retaining some extra reserves for small buybacks. If the market grows rapidly or the protocol develops at an accelerated pace (exceeding twice the target), full-scale buybacks are initiated. The buyback design is fairly standard. The absolute buyback figures might not be very large for now, as the treasury accumulation period has been short. Aave also started discussing buybacks after accumulating satisfactory capital. The buyback effects will take time to show. This proposal focuses more on formally allocating protocol revenue to risk reserves, essentially adding an insurance layer to the protocol.