Project Introduction
- Why it matters
- After Ethereum’s shift to Proof-of-Stake, two separate networking stacks keep the chain alive: LibP2P for the Consensus Layer (CL) and devp2p for the Execution Layer (EL). Existing fuzzing efforts cover only one side at a time, leaving protocol “blind spots” that attackers can exploit.
- What we will do
- Build the first unified, state-aware fuzzing framework that targets both LibP2P (CL) and devp2p (EL) in one workflow, systematically mutating packet sequences across the full peer life-cycle to expose subtle logic and implementation flaws.
- Concrete outcomes
- Open-source toolkit with plug-and-play support for ≥ 3 mainstream CL & EL clients and an optional CI pipeline.
- Academic publications, vulnerability advisories, and ready-to-merge patches that harden the Ethereum network layer.
- Early proof of traction
- Our 2024 Academic-Grant project delivered a time-series EL fuzzer, one patent filing, and a working prototype adopted by core-dev testers—establishing the engineering base we now extend to CL.
Code base
Our projects github is as following:
https://github.com/AgnopraxLab/D2PFuzz
Bug We found
https://github.com/ethereum/go-ethereum/issues/31471
https://github.com/paradigmxyz/reth/issues/15255
https://github.com/erigontech/erigon/issues/13931