After the Binance incident, researchers began asking whether rate limits and recovery should live in the asset (the token contract) rather than only in account code, and two ERC-20 limit proposals, ERC-8474 and ERC-8475, appeared on Magicians. potuz publicly questioned MetaMask and Rabby for defaulting to a priority fee close to 1 gwei. potuz also submitted EIP-8483 so validators can register post-quantum material, and EIP-8288 was proposed for compressing PQ signature data on the OP Stack.
Posted at 00:15 by dlebee. Smart wallets, ERC-7579 modules, social recovery and EIP-7702 delegation all put policy in account code, and that leaves two gaps. First, a contract that already holds a `transferFrom` approval can bypass the wallet's daily limit. Second, under 7702 a stolen EOA key can re-delegate and replace the original protection. Asset-level checks cover every path that reduces a balance, and in the long run the protocol could even apply them to native ETH.
ERC-8474 ERC-20 Delayed Unlock discussion thread and ERC-8475 ERC-20 Outgoing Window Limits discussion thread, both opened at 15:55, add delayed unlocks and time-window limits at the token level.
Posted at 13:48, breaks down the L1 transaction in which the Arbitrum Security Council froze 8,425 ETH and about 74.56 million ARB on addresses linked to the Binance incident on October 30: a 9-of-12 Safe called the UpgradeExecutor, temporarily swapped the Inbox implementation and sent the message.
05:23: wallets default to a priority fee close to 1 gwei when a tip measured in wei would do. Scaling makes transactions cheaper, but if half the fee is the tip, the scaling work is wasted.
Opened at 22:16 by potuz. It adds an EIP-7685 request type and a predeployed contract so a validator's withdrawal address can register 32 bytes of PQ material in the beacon state, preparing for a future PQ migration fork. The EIP-8483 discussion thread opened at 22:52.
EIPs PR #12609: Add EIP-8483 Post-Quantum Material Registration, opened at 22:16 by potuz. It adds an EIP-7685 request type and a predeployed contract so a validator's withdrawal address can register 32 bytes of PQ material in the beacon state, preparing for a future PQ migration fork. The EIP-8483 discussion thread opened at 22:52.
Alongside it, EIPs PR #12608: EIP-8148 request type moves to 0x06 (EIP-8148 is custom validator withdrawal thresholds) to avoid clashing with the new request type.
Posted at 09:05. If every L2 user switched to PQ signatures, signatures would take 99% of transaction data and the transfers that fit in 6 blobs would fall from about 6,600 to about 155. With EIP-8288 (mempool signature and proof aggregation), signatures move out of the transaction and a single leanVM aggregate proof goes onchain, bringing a transaction signed with leanSPHINCS down to 100 to 129 bytes.
Opened at 22:54, together with EIPs PR #12611: EIP-7928 static context stops before pre-state checks and EIPs PR #12613: EIP-7708 covers contract creation transactions and reverted frames. All are spec fixes for Glamsterdam content and do not change how anything is used.
EIPs PR #12612: EIP-8038 maps the EIP-7928 pre-state gas table to new constants, opened at 22:54, together with EIPs PR #12611: EIP-7928 static context stops before pre-state checks and EIPs PR #12613: EIP-7708 covers contract creation transactions and reverted frames. All are spec fixes for Glamsterdam content and do not change how anything is used.
EIPs PR #12612: EIP-8038 maps the EIP-7928 pre-state gas table to new constants, opened at 22:54, together with EIPs PR #12611: EIP-7928 static context stops before pre-state checks and EIPs PR #12613: EIP-7708 covers contract creation transactions and reverted frames. All are spec fixes for Glamsterdam content and do not change how anything is used.
22:47, explains what can go wrong when a bridge releases funds on one side as soon as it sees a deposit on the other, and previews their tooling for it.
