New primitive in efficient on-chain quantum resistant signature verification using the Falcon cryptography scheme.
Moosh.gg team members entry for Starknet re{ignite} hackathon 2025. We are entering for the 'Best use of Starknet' hackathon track.
Quantum-resistant signatures significantly increase on-chain storage requirements because they use much larger signature and public key sizes compared to traditional cryptography. This leads to higher gas costs for smart contract storage, reduced throughput (no. tx per block), and faster blockchain state growth (bloat). As a result, adopting post-quantum cryptography poses scalability and efficiency challenges for blockchains and Web3 Apps.
We propose a solution where a non-negligible tx gas expense is only required once, then further verification is gasless. This reduces smart contract storage requirements, increases tx throughput for any Starknet Web3 app or system requiring quantum resistant verification, and minimizes ongoing gas fees after the initial one-time setup.
For this submission demo we have applied the solution in the example use case of a generic on-chain marketplace, where person A (a client) pays Person B (a provider) for services/resources in an escrow smart contract with a dispute resolution mechanism. Our vision of a further real life application can be an on-chain marketplace where provided services are leases for peer-to-peer sharing of virtual machines.
Why Starknet?
It offers floating point capabilities in Cairo, so we can perform Falcon based Post-Quantum-Cryptography (PQC) signing
It offers Class based deployment which reduces fees if multiple contracts are deployed of the same class (We use for Escrow contracts). It also provides a mechanism to spot for malicious contract deployment.
Speed of it's blockchain - suitable for marketplaces which will have lots of network traffic.
In future, as Cairo is a ZK provable language, we can can use offline verification for Falcon signatures as an extension of this concept.
Presentation slides can be found at:
https://www.moosh.gg/moosh_files/starknetIgniteSubmissionMay2025.pdf
Mainnet contract classes:
Registry : 0x022a351ab5f1ac13352a3792be246d8c6513d9029f3674608ac4cd7944aa702e
Verifier : 0x0396507525f71d979d306df5b72c568dfcca173158086d73806e496b054670a3
Escrow : 0x0028172888cc58dece1ccaaadcd0b8076eb85f0284f95aecd28027042b0f64a9
This was built 100% during the hackathon time window. No work done prior to the event. We have built this version of the ID system concept to be open source and plan to develop it further for our own use cases in the future as we continue to explore deeper integrations with Starknet and cloud infrastructure.
N/A: we contribute to a pre-seed funded Web3 project, but this submission is not part of a fundraise.