LayerZero OFT DVN config (count, threshold, diversity)
A cross-chain & bridge factor in the v1.7.0 rubric. Measured per protocol on a s cadence.
Methodology how we score#
What this measures
This factor assesses the LayerZero OFT (Omnichain Fungible Token) adapter's DVN (Decentralized Verifier Network) configuration: how many DVNs are required to attest a cross-chain message, what threshold of those DVNs must agree (k-of-N), and whether the listed DVN operators are independent entities or the same operator running multiple endpoints. Cat 10 applies only to bridge-touching protocols; non-bridge protocols show this factor as N/A. The 1-of-1 DVN configuration is the catastrophic edge — a single DVN can unilaterally attest to a forged inbound message and trigger an unbacked mint on the destination chain.
Why it matters
LayerZero OFT is now the dominant cross-chain token-bridging primitive on EVM L2s, but its security guarantees are a function of how the OFT adapter is configured by the deploying protocol — not an intrinsic property of LayerZero itself. A 1/1 DVN configuration means there is no honest-majority assumption: a single compromised, captured, or buggy DVN can forge an `lzReceive` payload and the destination chain will accept the mint as legitimate. Threshold without operator diversity is similarly weak: 2-of-3 DVNs all run by the same operator collapses to single-operator trust at validator level, even though the on-chain config looks redundant.
Green / Yellow / Red
Green is ≥3 independent DVN operators with threshold ≥2 — a meaningful k-of-N attestation set. Yellow is 2 DVNs with threshold 2, or any threshold-2 configuration where operators are not fully independent. Red is the 1/1 configuration: a single DVN with threshold 1, where any single attestation suffices to mint on the destination chain.
Common gray cases
The protocol does not deploy a LayerZero OFT adapter at all (factor is N/A). On-chain DVN registry reads are inconclusive when the adapter uses a custom DVN list that is not standard-registered, requiring source inspection.
Notable historical examples
No cross-hacked incidents currently linked in database for this factor. The most-cited industry incident is Kelp DAO (Apr 2026, $292M loss), where a 1-of-1 DVN configuration on the rsETH OFT adapter allowed a single forged message to mint unbacked rsETH on a destination chain. Aave governance forum publicly flagged the 1/1 DVN risk approximately 15 months prior to the exploit, making this a structural risk that was visible on-chain and publicly debated long before failure.
Measurement what to look for#
For any LayerZero OFT adapter, read the DVN configuration: count of DVNs, k-of-N threshold, and operator diversity (independent operators vs same-operator multi-DVN).