LayerZero
TEMPEREDCross-Chain Messaging · Multi-chain · N/A (infra) TVL · 15 contracts
Public risk assessment — scores are produced with the same methodology as monitored protocols
Security Profile
70
65
60
52
55
82
64
72
55
75
84
70
65
60
52
55
82
64
72
55
75
84
Audit History
Bug Bounty Program
Assessment
Mature cross-chain messaging protocol with strong operational history but extreme compositional risk (D4=52, 100+ dependent protocols) and cross-chain trust assumptions (D10=55, DVN honesty model). Any LZ core bug cascades to entire ecosystem. Governance centralization (D5=55) and DVN trust model drag score down from DAMASCUS. Good maturity (D6=82) and adversarial resilience (all findings FP) prevent drop to FORGED.
Dimension Breakdown
How scores work →- 93 access control checks across 511 total checks (18.2% density)
- Complex cross-chain authorization model with endpoint-library separation
- onlyOwner (12 instances), onlyEndpoint, validVersion modifiers
- Graph extraction missed custom patterns (onlyTreasury, nativeFees[msg.sender])
- Admin can configure default libraries, DVNs, adapter params
- Cross-chain gas pricing model adds economic complexity
- Fee model across chains creates arbitrage surface
- Treasury fee accumulation (treasuryZROFees, nativeFees mappings)
- 137 state writes with fee-related writes prominent
- 5 price_feed edges in graph topology
- DVN replaces oracle model from V1 but adds trust assumptions
- DVN trust varies per pathway and configuration
- Default DVN set controlled by LayerZero Labs (centralization)
- hashLookup mapping is the verification state -- 4-deep nested mapping
- FPValidator adds proof verification layer
- LayerZero Labs retains significant control over core infrastructure
- DVN selection initially centralized (Google Cloud, Polyhedra default set)
- ZRO token governance immature
- registerLibrary, setDefaultSendVersion, setDefaultReceiveVersion all admin-gated
- Security council provides some decentralization
- V1 live since 2022, V2 since 2024 (~4+ years org maturity)
- No major exploits on core messaging infrastructure
- V1->V2 migration demonstrates architectural iteration
- Extensive audit coverage (Trail of Bits, Zellic, Code4rena)
- Z-factor: 0.847
- Score derived from continuous adversarial security research
- DVN trust model: message verification depends on DVN set honesty
- Message ordering: no guaranteed ordering across channels
- Replay protection implemented but cross-chain state sync inherently fragile
- Liveness: DVN failure can halt message delivery per pathway
- 30+ chain deployments: each chain adds verification surface
- 113 reentry edges -- cross-chain callbacks create reentrancy surface
- 100+ protocols depend on LayerZero for cross-chain messaging
- Any core vulnerability cascades to entire ecosystem
- 103 external calls across 9 contracts
- OApp integration bugs are outside LayerZero control
- Stargate, Radiant, Aptos bridge all depend on LZ
- Custom messaging libraries (non-standard patterns)
- Complex dependency graph across chain deployments
- OFT standard adds integration complexity
- 4 trust_dependency edges in core graph
- Dual ULN versions (301, 302) add maintenance surface
- Active cross-chain monitoring infrastructure
- Security council with incident response capability
- Bug bounty program active on Immunefi
- Demonstrated response during V1->V2 migration
- Appears in 1 cross-protocol cascade chain(s)
- Failure cascades to 4 downstream protocol(s)
- Member of 1 dependency cluster(s)
- Score: 84/100 (higher = more isolated from systemic risk)
- Source: cross_protocol_composition.json dependency analysis
Risk Drivers
Primary risk factors driving this score, ordered by severity.
Adversarial Risk Signals
Observable security posture indicators. These signals reflect publicly verifiable information and responsible disclosure outcomes. No specific vulnerability details are exposed.
Score History & Verification
Score provenance tracking begins with the next reassessment.
On-Chain Data
- Protocol Slug
- "layerzero"
- Oracle
- BRORegistry (Base)
- Evidence
- IPFS (pinned)
- Staleness Threshold
- 24 hours
registry.getScore("layerzero")Reduce exploitable risk
BlackHart Monitoring provides continuous adversarial analysis, vulnerability detection, remediation support, and verified reassessment when your risk posture improves.