Search for projects by name
Cartesi PRT Honeypot v2 is an application-specific Stage-2 rollup that stress-tests Cartesi Rollups’ security. Protected solely by Cartesi’s PRT (Permissionless Refereed Tournaments) fraud-proof algorithm, it turns its locked funds into an open bounty for... anyone who can break the system. Users should not deposit unless they are willing to donate their funds to the Honeypot.
Tokens breakdown
Cartesi PRT Honeypot v2 is an application-specific Stage-2 rollup that stress-tests Cartesi Rollups’ security. Protected solely by Cartesi’s PRT (Permissionless Refereed Tournaments) fraud-proof algorithm, it turns its locked funds into an open bounty for... anyone who can break the system. Users should not deposit unless they are willing to donate their funds to the Honeypot.
The section shows the operating costs that L2s pay to Ethereum.
2025 Nov 12 — Dec 03
Cartesi PRT Honeypot v2 deployed on Ethereum
2025 Nov 8th
Cartesi PRT Honeypot v2 deployed to Ethereum mainnet.
Dave: decentralized, secure, lively fraud-proofs
2025 May 9th
Dave paper published on ACM DLT, a peer-reviewed journal.
Fraud proofs allow actors watching the chain to prove that the state is incorrect. Interactive proofs (INT) require multiple transactions over time to resolve.
All of the data needed for proof construction is published on Ethereum L1.
Users cannot exit their funds as all deposits are considered donations.
Rollup operators cannot compromise the system, but being application-specific might bring additional risk.
Users can deposit (donate) CTSI tokens to the Honeypot. The funds can only be withdrawn by the Cartesi Multisig to its own address. The appchain has the very specific purpose of a bug bounty on the proof system, incentivizing security researchers to break it and claim the deposited funds.
The genesis state comes from the Honeypot Cartesi Machine template included in the Honeypot v3 release. Alternatively, you can recreate it by following the build steps in the Honeypot GitHub Repository.
The information in the section might be incomplete or outdated.
The L2BEAT Team is working to research & validate the content before publishing.
After some period of time, the published state root is assumed to be correct. For a certain time period, usually one week, anyone can submit a fraud proof that shows that the state was incorrect.The initial bond for joining the tournament is set to 0.23219805 ETH.
Funds can be stolen if there is no one that checks the published state. Fraud proofs assume at least one honest and able validator.
There is no privileged entity that sequences transactions or produces blocks. This activity is permissionless and open to anyone.
Because the state of the system is based on transactions submitted on the underlying host chain and anyone can submit their transactions there it allows the users to circumvent censorship by interacting with the smart contract on the host chain directly.
No address apart from the Cartesi Multisig can trigger a withdrawal and deposits are considered donations to the Honeypot. If a withdrawal is requested by them, all funds in the escrow are withdrawable to the permissioned address as soon as the next settlement on L1 occurs (min. every 7d 1h).

A Multisig with 3/6 threshold.


Serves as both the canonical log for arbitrary dApp inputs and a portal for depositing assets (one possible type of input). It ensures data availability and that all off-chain participants process the same inputs in the same order.
Contract managing PRT fraud-proof tournaments, application epochs and input validation, as well as settlement and challenge periods. Dispute tournaments are started here and the final, verified computation result (as an outputsMerkleRoot) is recorded when they are resolved.
Main dApp contract that escrows assets and executes the verified results (outputs) from off-chain computation. It relies on the DaveConsensus contract to validate outputs before releasing assets or triggering on-chain actions. The immutable template hash of the dApp is 0x144d45af1181b35f2b11c4b1150d6cb16934c28093707fb97c911ff16b3fe609.
Handles the intermediate stages of a dispute following the TopTournament targeting a more granular bisection game.
Onchain verifier that can execute a single, disputed instruction of the Cartesi machine. It is the ultimate arbiter that BottomTournament calls to determine which party’s claimed state transition is correct.
Responsible for creating and orchestrating the multi-stage dispute process. It instantiates the correct tournament contract (Top, Middle, or Bottom) depending on the current stage of the dispute game.
Represents the entry point and highest level of a dispute in PRT. Disagreeing validators join this tournament to resolve conflicts over the entire computation trace through a bisection game. The required bond amount for joining the Tournament is calculated from worst case gas estimates and currently is 0.23219805 ETH.
Contract that allows anyone to perform transfers of ERC-20 tokens to Cartesi DApps.
Provides constant configuration data for the tournament system. It defines parameters like the number of levels (3), the minimum challenge period of ~7d, and the size of computation segments at each stage of a dispute.
Referees the dispute over a single contested Cartesi machine step as the final stage of arbitration in a dispute. It calls the CartesiStateTransition contract to get a definitive on-chain ruling and identify the winner.
Contract storing bounty funds.