Dinero Bounty Details

• Until TVL > $5,000,000
  • Max Critical: $500,000 in ETH
  • High Severity: $30,000 in ETH
• Once TVL > $5,000,000
  • Max Critical: 10% of impacted TVL, up to $2,000,000 in ETH
  • High severity: $50,000 in ETH
• Join C4 Discord to register
• Submit findings using the C4 form
• Read our Code4rena Blue guidelines for more details

❗ Note for participants: The sponsor's repo, scope definition, and contents herein are all subject to change.

Publicly Known Issues

• All issues submitted via wardens and the Blue Team during this Code4rena Blue engagement will be added to this project list weekly.

• Centralization Risks: Some methods (such as emergencyWithdraw) are only accessible by the Redacted DAO multisig, which is the sole owner of the contracts. This is acceptable as the multisig is controlled by the Redacted DAO, which is a decentralized organization. These methods would only be used for emergency purposes, such as in the event of a critical bug or a hack.

• ERC-1155 Mint Re-entrancy: The contract is not vulnerable to reentrancy attacks because the contract does not hold any funds and does not call any external contracts in the same transaction as the mint call. The contract is also not vulnerable to re-entrancy attacks because the contract does not use any state variables that can be modified by an external contract in the same transaction as the mint call.

• Deposit Front Running: This issue is mitigated by pushing all validators into queue via access control once they have an effective balance of 1 ETH, meaning they have already been registered with the canonical beacon chain deposit contract.

• Allowances and OPERATOR_ROLE: The OPERATOR_ROLE is able to set allowances for pxETH. This role only given to the PirexETH contract and is used to facilitate fee distribution. The OPERATOR_ROLE is not given to any external contracts or accounts.

• Effects of setContract on State: Changing withdrawal credentials aka the rewardRecipient contract address could corrupt state. For example, if there are initialised validators and rewardRecipient is changed via setContract, then functions like getInitializedValidatorAt may return incorrect withdrawal credentials. This is mitigated by the fact that setContract can ony be called by the owner (Redacted DAO multisig) which does extensive due diligence before executing any transactions, and rewardRecipient is not expected to change.

Note: We have acknowledged all findings in referenced Audits and have either fixed them or have mitigated them. These functions are required for the protocol to work as intended.

Pirex ETH Overview

Pirex ETH is built on top of the Redacted DAO’s Pirex platform and forms the foundation of the Dinero protocol. It is a two-token system built around ETH staking, consisting of pxETH and apxETH, tailored for different user preferences. This design gives users a choice: pxETH for liquidity or apxETH for boosted ETH staking yield.

pxETH and apxETH

When depositing ETH, users can choose between holding pxETH or depositing to an auto compounding rewards vault for apxETH.

• pxETH is for those willing to forgo staking yield for liquidity. When users choose to hold pxETH, they’re opting to hold an ETH-pegged asset that can take advantage of opportunities throughout DeFi. These include providing liquidity, participating in lending protocols, and more. The Redacted DAO will be using its treasury and BTRFLY incentives to expand such opportunities for pxETH holders.

• apxETH is for users focused on maximizing their staking yields. After minting pxETH, users can deposit to Dinero's auto-compounding rewards vault to enjoy boosted staking yields without the hassle of running their own validators. Since some users will choose to hold pxETH, each apxETH benefits from staking rewards from more than one staked ETH, amplifying the yield for apxETH users.

Deposits and the ETH Buffer

Most ETH deposited into the Dinero protocol via Pirex ETH is staked on the Ethereum network. However, a small fraction remains in an "ETH buffer" instead of being staked. This buffer facilitates smooth staking and unstaking, allows faster ETH withdrawals when it has funds, and will support self-contained meta transactions through the Redacted Relayer RPC in the future.

Withdrawals

There are limits on the rate at which validators can enter and exit the Ethereum network, based on the total number of validators. Therefore, if there is a significant ETH unstaking queue, this can hamper the timeliness of ETH withdrawals from the Dinero protocol from the spinning down of validators. In these circumstances, an incentivized withdrawal pool can be used to improve pxETH liquidity from ETH unstaking.

Users can deposit ETH into a pool and receive rewards whilst they provide liquidity to that pool. Where there is an unstaking queue and ETH from the spinning down of validators is not readily available, ETH from this pool is provided to users in exchange for pxETH, with the exchange rate or price being determined by demand for ETH from the pool. As pxETH is redeemed and validators are spun down, ETH is replenished in the pool. Depositors into the withdrawal pool therefore receive rewards in exchange for potential ETH illiquidity.

As pricing depends on the demand for ETH in the pool, rewards on deposited ETH increase during periods of high demand, allowing the pool to scale when demand is high. This makes liquidity provision more efficient and cost effective.

Yield Stripping (Coming Soon)

Yield from apxETH can be tokenized via yield stripping. For example, if a user wants to tokenize 1 year of yield for 1 pxETH deposited in the rewards vault, they can exchange 1 pxETH for:

• 1 pxETH principal semi-fungible token which can be exchanged for 1 pxETH in one year; and
• 1 pxETH yield semi-fungible token for each rewards period in the next year, which can be exchanged for the rewards earned by 1 pxETH in the rewards vault in one rewards period.

Users decide how many reward cycles they tokenize. These tokens can be used throughout DeFi and are tradable. Yield stripping provides users the ability to leverage, hedge, and speculate on future pxETH price and future yield.

Links

• Previous audits:
  • Spearbit - PirexETH (@spearbitdao)
  • Pashov - PirexETH (@pashovkrum)
• Documentation:
  • PirexETH & Dinero Documentation
  • PirexETH Whitepaper
  • Dinero Litepaper
• Website: https://dineroismoney.com
• Twitter: (@redactedcartel)
• Discord: https://discord.gg/redactedcartel

Scope

This is the complete list of what's in scope for this contest:

This is a list of mainnet contract deployments:

Out of scope

Contracts:

• ChainlinkFunctionsOracleAdapter.sol

Vendor Libraries:

• chainlink
• solidty-cborutils
• @ensdomains/buffer

Additional Context

• Trusted Roles

  • DEFAULT_ADMIN_ROLE: Set external contract addresses
  • GOVERNANCE_ROLE: Manages validator queue, set fee params, pause deposits to beacon chain deposit contract
  • KEEPER_ROLE: Harvest rewards, update status when a validator is slashed and top up validator stake when active balance goes below effective balance
  • OPERATOR_ROLE: Allows PirexETH to perform specific internal actions (eg. approve allowances for fee distribution)
  • ORACLE_ROLE: Updates the state of the validator when it is dissolved
  • MINTER_ROLE: Allows PirexETH to mint pxETH and apxETH
  • BURNER_ROLE: Allows PirexETH to burn pxETH and apxETH

• EIP Specifications:

  • pxETH: Should comply with ERC-20 standard
  • apxEth: Should comply with ERC-4626 standard
  • upxETH (Unlocking) and RFN (Yield Stripping): Should comply with ERC-1155 standard

• In the event of DOS, we would consider a finding to be valid if it is reproducible for a minimum duration of 4 hours.

Main invariants

• Setting and updating contract addresses (pxETH address, etc) which are controlled by the DEFAULT_ADMIN_ROLE
• pxETH and apxETH are minted and burned in a 1:1 ratio
• outstandingRedemptions
• pendingWithdrawal
• pendingDeposits

Scoping Details

• If you have a public code repo, please share it here: https://github.com/redacted-cartel/pirex-eth-contracts
• How many contracts are in scope?: 17
• Total SLoC for these contracts?: 3519
• How many external imports are there?: 21
• How many separate interfaces and struct definitions are there for the contracts within scope?: 5 interfaces & 3 structs
• Does most of your code generally use composition or inheritance?: Yes, inheritance
• How many external calls?: 1 - Beacon Chain Deposit Contract
• What is the overall line coverage percentage provided by your tests?: 99.8%
• Is this an upgrade of an existing system?: No
• Check all that apply (e.g. timelock, NFT, AMM, ERC20, rollups, etc.): ERC-20, ERC-1155, ERC-4626
• Is there a need to understand a separate part of the codebase / get context in order to audit this part of the protocol?:
• Please describe required context:
• Does it use an oracle?: Yes - Chainlink for triggering dissolve validator
• Does it use a side-chain?: No
• Describe any specific areas you would like addressed:

Miscellaneous

Employees of Redacted Finance, and employees' family members are ineligible for bounties.