Mantle Bridge

Rejected venues wait the longest for re-review; a rejection has to earn another look before the scheduled date.

REJECTED ON THE SAME AXES THAT SANK WORLD CHAIN AND LINEA IN THIS BATCH. Mantle Network has made real architectural progress — an OP-Succinct upgrade in September 2025 replaced optimistic fraud-proof validation with SP1 zkVM validity proofs, and an April 2026 upgrade moved data availability to native Ethereum blob space — but L2Beat rates its exit window as ”None,” the identical rating already used to reject Linea in this registry, because contracts are instantly upgradable through a 6-of-14 multisig with no delay on any material contract except the token itself. State-root proposers are permissioned, not open, paralleling the permissioned-validator structures already rejected for Robinhood Chain and World Chain. Real mechanism-level progress does not change the governance-concentration facts this registry actually prices.

Kill criteria: the events that revoke this verdict
• State-root proposer verification becomes permissionless, removing the current whitelist requirement
• A minimum standing delay, even 24 to 72 hours, is introduced between MantleSecurityMultisig upgrade approval and execution
• L2Beat no longer rates Mantle’s exit window as ”None”
• Current withdrawal timing is independently confirmed against a live transaction rather than conflicting documentation

The research file

Mechanism

Mantle Network is an OP-Stack-derived rollup. Two recent architecture changes are material: the OP-Succinct upgrade of 2025-09-16 replaced optimistic fraud-proof validation with SP1 zkVM validity proofs, and the Arsia upgrade of April 2026 moved data availability from EigenDA to native Ethereum blob space. Sources conflict on current withdrawal timing: Mantle’s own bridging documentation still describes a seven-day rollup challenge period, while independent reporting states the ZK upgrade cut withdrawals to roughly six hours; this review could not resolve the discrepancy against a live transaction, which is itself a documentation-quality concern independent of the governance findings below.

Control and governance

Confirmed directly from L2Beat: Mantle’s exit window is rated ”None,” identical to the already-rejected Linea. Most contracts upgrade through a 6-of-14 MantleSecurityMultisig with no delay; only the L1MantleToken itself carries a one-day timelock. State-root proposers are permissioned, whitelisted only, paralleling the already-rejected Robinhood Chain and World Chain’s permissioned-validator structures. The sequencer is centralized, with a 12-hour force-inclusion fallback for censored transactions. L2Beat rates Mantle Stage 0, five of eight requirements met, three unresolved for Stage 1, and flags instant contract upgrades as a CRITICAL risk alongside centralized-validator and permissioned-proposer risks — the same critical-risk pattern that sank World Chain and Robinhood Chain in this batch.

Incident record

No exploit of Mantle’s own bridge or core contracts was identified. Mantle publicly clarified on 2025-04-15 that the Kelp DAO/LayerZero rsETH exploit already rejected in this registry’s `layerzero-v2` entry did not touch Mantle’s official bridge or core infrastructure, even though Mantle was one of over 20 chains where LayerZero-bridged rsETH circulated — a dependency-adjacent near-miss rather than a Mantle-specific failure.

Exit under stress

Whatever the current numeric withdrawal window turns out to be, the more decisive fact is the same one that sank Linea: zero delay on contract upgrades means there is no exit window at all if the 6-of-14 multisig approves a malicious or coerced change, regardless of how fast normal-path withdrawals otherwise are.

Comparison

Weaker than Base, Arbitrum, and Optimism, all approved with limits in this registry, on the same two axes that already disqualified World Chain, Robinhood Chain, and Linea: a permissioned rather than permissionless validator or proposer set, and an upgrade path with no standing delay, rated ”None” by L2Beat exactly like Linea’s. Mantle’s move to ZK validity proofs and Ethereum-blob data availability is a genuine mechanism-level improvement, but it does not touch the governance-concentration facts that are the actual basis for this rejection.

Sources

The claims above trace to these. Where a number could not be independently verified, the thesis says so.

• L2Beat — Mantle · primary · accessed 2026-08-17
  Supports: exit window: None, 6-of-14 MantleSecurityMultisig, permissioned proposer, Stage 0, critical risk flags
• Mantle Network — bridging on Mantle mainnet · primary · accessed 2026-08-17
  Supports: deposit and withdrawal flow
• Coin Bureau — Mantle Network review 2026 · secondary · accessed 2026-08-17
  Supports: OP-Succinct upgrade, Arsia DA migration, reported withdrawal timing
• Cryptonews — Mantle Network stands resilient, core infrastructure unscathed by Kelp DAO breach · secondary · accessed 2026-08-17
  Supports: Mantle statement distinguishing its bridge from the LayerZero exploit
• Mantle Bridge FAQ · primary · accessed 2026-08-17
  Supports: withdrawal FAQ

Inherited controls

The verdict above grades the protocol layer. Every position also inherits the asset it holds and the chain it settles on. The least safe layer sets the position’s grade, and the position table names which one that is.