This is an archive of our post on Aave governance forum. Read the full thread here.

Summary

LlamaRisk supports deployment to the Mantle network with specific considerations. While Mantle shows strong growth, its reliance on 10 permissioned data availability nodes and incomplete fault-proof system (earning Alt-DA classification) requires attention. This reliance raises centralization concerns, though the network is actively working to address these. The March 2024 Tectonic upgrade brought improvements through EIP-1559 implementation and 2-second block intervals.

A gap in the security infrastructure is the absence of a bug bounty program. We strongly recommend implementing this imminently to align with security best practices across other major networks.

The market infrastructure is robust, with established DEXes (Merchant Moe, Agni Finance) and lending protocols (INIT Capital), multiple oracle solutions, and comprehensive bridging. Research shows that Aave deployments contribute to network growth by attracting capital and fostering liquidity distribution across protocols over time. The Treasury’s substantial holdings ($4B, predominantly in MNT) and active grants program suggest strong growth potential. However, centralized components for critical functions (Sequencer, Proposer, Challenger) warrant monitoring.

We support deployment with conservative initial parameters that can be adjusted as the network matures, decentralizes, and implements key security measures like the bug bounty program.

Our analysys

1. Network Fundamental Characteristics

1.1 Network Overview

The Mantle Network is a modular L2 scaling solution built atop Ethereum, leveraging Optimistic Rollup technology. The integration of EigenDA ensures extended transaction data storage and retrieval capabilities. This modular approach to data availability is expected to yield cost efficiencies with over 90% reduction in fees compared to traditional L1s.

The implementation of the Mantle v2 Tectonic upgrade in March 2024 marked a significant architectural advancement. This upgrade removed redundant components like the Data Transport Layer and Threshold Signature Scheme, resulting in a more efficient infrastructure. The Tectonic upgrade also migrated Mantle’s codebase from OVM to OP Stack. Notable enhancements include EIP-1559 implementation for improved price auction dynamics and upgraded L2 block generation that maintains consistent 2-second intervals, enabling efficient transaction batching.

Key distinctions between Mantle v2 and OP Stack Bedrock stem from Mantle’s independent data availability solutions, native token migration, and optimized fee mechanics.

Performance metrics demonstrate Mantle Mainnet’s operational efficiency, with an average block generation time of 0.639 seconds. The network has processed 184.9M transactions across 73.666M blocks, with a TPS averaging 4.25, which is approximately one-third of Ethereum’s throughput. Transaction costs remain stable between 0.01-0.03 MNT, maintaining 92-98% processing success rates, providing reassurance about the network’s performance.

Source: Mantle Explorer, December 30th, 2024

Architecture

Network architecture is designed around the ability to interact with EigenDA-supported Mantle Data Availability to store Rollup data and publish updated state data to the Ethereum L1.

Source: Mantle Docs, December 30th, 2024

Below are described the key components and their functions:

• Sequencer: Processes incoming transactions and orders them into blocks. It comprises op-geth (a customized version of Ethereum’s Go client for execution) and op-node that handles rollup-specific logic, ensuring compatibility with L1.

• Proposer: After processing, the Sequencer forwards state updates to the op-proposer responsible for initializing and confirming state updates. op-proposer fetches the latest state root from Ethereum L1 and submits updated state roots to ensure consistency.

• MantleDA: Stores transaction data off-chain securely and fetches data frames as required. Works closely with the op-batcher to ensure data integrity and accessibility.

• Batcher: Fetches blocks from the Sequencer and compiles them into data frames for submission to the MantleDA. Ensures data frames are properly prepared for verifier synchronization.

• Verifier: Ensures the validity of the processed data by synchronizing block and data status. Has a similar composition to the Sequencer.

Independent node operators can take none of these roles as of the date of the report. Nodes are not yet open for security reasons, which limits the operation to the core team-controlled or selected entities.

MantleDA

Mantle DA is a fork of EigenDA, adapted to solve specific needs of the Mantle network. A permissioned set of nodes (currently 10, corresponding addresses are indicated in the docs) receive the transaction payload, sign it using the BLS signature scheme and return the signatures to the Sequencer. The Sequencer then aggregates these signatures and posts the commitments to Ethereum’s DataLayrServiceManager (DA Bridge) contract. DataLayrServiceManager ensures that the signatures are valid and originate from node operators who are part of the agreed quorum (9/10).

To join the network, node operators must stake at least 100,000 MNT. Registration is managed by Register Operator Manager, who is authorized to register or change operator status. MantleDA lacks a slashing mechanism to penalize misbehaving nodes.

DataLayrServiceManager contract: owned by MantleEngineeringMultisig
BLSRegistry contract: owned by MantleEngineeringMultisig

L2Beat categorizes Mantle as a universal-purpose Optimium, placing it alongside Validiums and other Alt-DAs/light Layer 2s solutions. Unlike Rollups, which post all data directly to Ethereum, Alt-DAs periodically submit state commitments to Ethereum for validation but omit to post the underlying data.

Due to external trust assumptions to mitigate data withholding risks, Mantle has been flagged for transition to the “Other” category on L2Beat. This classification highlights the network’s current lack of an implemented proof system and the absence of data availability guarantees. A reassessment of Mantle’s categorization is expected within approximately 160 days. If the network demonstrates significant advancements in its architecture, including improvements to its proof system and data availability mechanisms, it could regain its status as a Stage 0 or higher-ranked solution.

Security

Mantle’s codebase has been audited by Secure3, OpenZeppelin, and Sigma Prime. This security verification builds upon the extensively audited OP stack, which forms the foundational architecture of Mantle’s protocol.

• Mantle_V2_ Secure3 Audit Report (March 2024) - 8 critical, 29 medium, 17 low severity issues;

• Mantle Node, Batcher, Proposer, and Tooling Incremental Final Audit Report (March 2024) - 1 medium, 9 low severity issues;

• Mantle Op-Geth Audit Final Report (March 2024) - 3 critical, 5 low severity issues;

• Mantle V2 Solidity Contracts Audit Report (March 2024) - 1 critical, 3 medium, 7 low severity issues;

• Sigma_Prime_Mantle_L2_Rollup_V2_Security_Assessment_Report_v2_0 (April 2024) - 4 critical, 4 high, 4 medium, 5 low severity issues;

The network maintains full transparency through its open-source GitHub repository, providing stakeholders with the opportunity to stay informed and involved. Development activity was intense throughout 2021-2023, with a notable resurgence observed in early 2024, primarily attributed to the implementation and refinement of the v2 Tectonic upgrade.

Source: Mantle Github, December 30th, 2024

There is no bug bounty for the Mantle network. An active program on Immunefi covers only Mantle LSP.

Fees

Mantle v2 fees are calculated through a dual mechanism consisting of the L2 Execution Fee and the L1 Rollup Fee.

The L2 Execution Fee refers to the expense incurred for processing and executing transactions directly on Mantle. The formula determines the Execution Fee:

L2gasPrice depends on transaction type (EIP-1559 or legacy), and the gas used reflects transaction complexity. Gas on the Mantle is denominated and payable in $MNT.

The L1 Rollup Fee represents the cost of submitting the rollup’s state root to L1. This fee is evenly distributed among the users whose transactions are included in a specific batch. It covers the expense of recording the rollup data and state credentials onto the DA contract on L1. Rollup Fee is calculated as:

rollupDataGas is based on the size of the calldata, while overhead/scalar values are retrieved from BVM_GasPriceOracle.

Fee structure in Mantle v2 relies on tokenRatio - a dynamic value calculated as eth_price/mnt_price, adjustable via BVM_GasPriceOracle by authorized users. The Oracle contract has EOAs as owner and operator. According to the documentation, updates to the token ratio are intended to be executed via multisig with HSM. However, tokenRatio adjustments are being directly carried out by the operator at this review.

EigenDA utilization for data storage allows complex L2 transactions to remain cost-efficient as only state roots and necessary data are sent to L1. EIP-1559 transactions involve a GasTipCap, a parameter for handling rewards to sequencers, which can be set to zero. In this order, the suggested transaction settings are baseFee = 0.02 gwei and priorityfee = 0. Users can directly calculate total transaction costs (L1 + L2) using the estimateGas API in Mantle v2.

1.2 Decentralization and Legal Evaluation

No disclosures regarding the entity(ies) operating https://www.mantle.xyz/ and connected front-ends have been made. Mantle team shared with us that they utilize a Cayman Islands-domiciled company to facilitate the execution of third-party agreements. Terms and Conditions stipulate that “Mantle Network is a layer-2 blockchain incubated by a decentralized autonomous organization”; therefore, the structure of Mantle DAO is to be examined further.

Mantle’s governance architecture predominantly relies on off-chain mechanisms. The process typically begins when the core development team or established contributors present strategic initiatives, which are then opened to community deliberation through dedicated forum discussions. When these discussions generate substantial community engagement and demonstrate strategic merit, they can evolve into formalized Mantle Improvement Proposals (MIPs), setting the stage for token-holder voting.

The protocol utilizes Snapshot as its primary voting infrastructure, implementing a voting mechanism where each delegated $MNT token corresponds to one vote. Mantle’s delegation system enables token holders to entrust their voting authority to experienced community representatives. The delegation isn’t merely an optional feature but is a fundamental prerequisite for proposal creation and voting participation. The proposal threshold is set at 200,000 $MNT while the quorum is 100,000,000 $MNT.

Access Control

Control functions are granted to the following addresses:

• MantleSecurityMultisig, Safe with 6/13 threshold;

• MantleEngineeringMultisig, Safe with 3/7 threshold;

• EOA1, a hot wallet;

• EOA2, a hot wallet;

Security and Engineering multisigs have some common signers.

• Challenger is specifically permitted to challenge or delete state roots proposed by the Proposer. Challenger’s role is tied to MantleEngineeringMultisig.

• Guardian can pause deposits and withdrawals in the system. It is assigned to MantleEngineeringMultisig.

• Proposer posts new state roots from the current Mantle layer to the host chain. It is controlled by EOA2.

• Sequencer is tasked with committing transactions from Mantle’s Layer 2 to the host chain. It is controlled by EOA1.

MantleSecurityMultisig has broader permissions; it can act on behalf of key system components, such as MantleTokenProxyAdmin and ProxyAdmin, enabling it to update system configuration, gas settings, and batch submitter addresses. Additionally, it can upgrade designated smart contract implementations, including BLSRegistry, L2OutputOracle, OptimismPortal, and L1StandardBridge.

Nodes documentation provides a high-level overview of the roles. Currently, only the Verifier function is accessible to node operators, while the other roles remain restricted. The limitation is attributed to the cited security concerns.

1.3 Activity Benchmarks

Mantle’s financial metrics reflect substantial market confidence, with bridged assets totaling $1.554B. Current TVL stands at $2.20 B, with a significant portion - $393.25M - concentrated in stablecoin positions. The network landscape shows diverse strength across multiple sectors: Pendle leads with $140.66M TVL, and decentralized exchanges showcase liquidity through Merchant Moe ($124.24M) and Agni Finance ($106.3M). The ecosystem’s financial infrastructure is further aided by Treehouse Protocol’s liquid staking ($97.49M TVL) and INIT Capital’s lending ($97.17M TVL).

Source: L2BEAT, January 6th, 2025

Daily active users trend shows significant fluctuations, with peaks reaching close to 200,000 users. New user registrations exhibit irregular bursts with a few prominent peaks.

Source: Dune.com, December 30th, 2024

Source: Dune.com, December 30th, 2024

2. Network Market Outlook

2.1 Market Infrastructure

Bridge

Native Mantle Bridge is the primary entry venue with broad support of networks - Ethereum Mainnet, Base, Optimism, Arbitrum, BNB Smart Chain, and Polygon. Additionally, there are third-party bridging solutions:

• Symbiosis

• RhinoFi

• Orbiter Finance

• Layerswap

Lending

Currently, the competition in the lending category is dominated by INIT Capital.

• INIT Capital ($97.37M TVL)

• MethLab ($22.05M TVL)

• Lendle ($20.39M TVL)

• Compound ($4.57M TVL)

• Minterest ($3.96M TVL)

• Dolomite ($3.49M TVL)

DEXes

Both decentralized exchanges on Mantle are native to the network (they’re not multichain present)

• Merchant Moe with TVL of $124.62M

• Agni Finance with TVL of $106.47M

Oracles

Pricefeeds on Mantle can be accessed through:

• Chainlink (CCIP only at this time, Data Feeds and Data Streams planned)

• API3

• Chronicle

• Pyth

• Redstone

• ORA

2.2 Liquidity Landscape

A lookup with Merchant Moe yielded the most liquid pools:

Source: Merchant Moe, December 30th, 2024

These rankings are confirmed with a liquidity check on Agni Finance, where cmETH sits on $45.95M TVL, followed by mETH ($27.15M TVL), WETH ($10.97M TVL), FBTC ($9.57M TVL), and USDe ($8.97M TVL)

Mantle Rewards Station is designed to incentivize and reward $MNT token holders. By locking their $MNT tokens for flexible or fixed periods, participants can earn rewards from various initiatives and protocols associated with Mantle. Active pools have included collaborations with projects like Ethena and mETH, offering earnings in $ENA and $COOK. The campaign has more than 30k participants who locked a total of 122.82M MNT (equal to 147.39M USD)

2.3 Ecosystem Resilience

While leveraging the OP stack provides Mantle with established infrastructure, it also inherits certain architectural limitations that warrant careful consideration. The key distinguishing factor in Mantle’s L2BEAT classification stems from specific challenges within its proof system architecture. The centralized nature of state management is of particular concern, where the protocol’s security model heavily depends on individual entities for state updates.

Other key functions, i.e., the ability to challenge or delete state roots proposed by a Proposer (Challenger role) and the power to pause deposits and withdrawals (Guardian role), are concentrated within MantleEngineeringMultisig.

At the date of the report, Mantle DA is supported by 10 permissioned operators. The system’s reliance on a limited number of attestors creates a scenario where collusion between these entities and the Proposer could lead to the finalization of an unavailable state. This critical vulnerability could compromise user funds.

2.4 Ecosystem Growth Potential

Mantle offers full EVM compatibility, making it easy for existing Ethereum-based applications to migrate or expand to the network. Integrating its native token, $MNT, into fee structures further reduces costs compared to other networks that rely on $ETH as their gas token, providing an economic incentive for users and builders.

Dominant assets within the network predominantly consist of BTC and ETH derivatives, with USDT absent from the top five by market capitalization. This dynamic creates a significant opportunity for money markets planning to launch on Mantle. These entities could differentiate themselves by incorporating a broader range of assets into their offerings and developing yield-generation competitive products tailored to meet the diverse demands of the network’s participants.

Grants

Mantle Grants Program has entered Season 2, spanning January to June 2025, focusing on fostering innovation in consumer, payments, and AI applications. It also aims to bootstrap utility for key assets within the ecosystem, including FBTC, mETH/cmETH, aUSD, and USDe.

At the program’s core is the Scout model, which enlists prominent figures from the web3 space, such as investors, founders, and key opinion leaders. These scouts are instrumental in sourcing and recommending promising early-stage projects. The grants are divided into three categories: early-stage grants, expansion grants, and venture support.

2.5 Major and Native Asset Outlook

Source: Coingecko, January 3rd, 2025

Except for MNT, Mantle’s ecosystem is dominated by ETH and BTC derivatives, constituting a significant share of its assets. Since November 2024, the network has experienced a notable surge in bridged TVL, climbing from over $1.5B to an all-time high of $1.6B in December 2024.

Among externally bridged assets, USDe stands out with a $101M in TVL, while canonically bridged tokens include MNT, mETH, ETH, and USDT, forming the core of the network’s liquidity.

2.6 Tokenomics

Beyond its governance functionality, $MNT is a fundamental component of Mantle Network’s economic model. The token is the primary medium for transaction fee settlement and the main incentive asset in Mantle Rewards Station.

The initial token distribution was outlined through two governance proposals - BIP-21 and MNT-22, establishing the groundwork for Mantle’s tokenomics.

Source: Mantle Docs, December 23rd, 2024

A significant portion of $MNT resides within the Mantle Treasury, representing tokens outside the circulating supply. The Treasury encompasses several strategic holdings, including core DAO-governed reserves, undistributed token allocations, and strategic DAO-owned liquidity pools. The deployment of $MNT from the Treasury requires formal Budget Proposals to receive community authorization before any distribution can occur.

Mantle Treasury has secured the third position in the DefiLlama rankings, trailing only the Ethereum Foundation and Golem Network. Over 90% of the Treasury’s value is held in MNT, with an estimated worth of $4.06B. Meanwhile, BTC and ETH assets contribute a combined $245.65M, and stablecoins account for $42.34M of the Treasury.

3. Onchain discoverability

Mantle maintains its proprietary blockchain explorer - Mainnet Explorer. Mantlescan is another block explorer and analytics platform built by Routescan. Network traceability is possible on TokenTerminal and Defillama. The ecosystem’s transparency is further enhanced through team-led and third-party-driven analytics on Dune.

4. Impact of AAVE Deployment

The primary risks to AAVE on Mantle stem from the network’s incomplete fault-proof system and limited decentralization of its DA layer. While upgrades to consensus mechanisms could have unintended consequences on user funds if implemented imperfectly, Mantle’s ecosystem has reached sufficient maturity to support DeFi operations.

The network has established key infrastructure elements necessary for AAVE deployment:

• Oracle Providers

• Robust bridging mechanisms

• Various liquidity venues

A significant consideration is the network’s liquidity depth. Limited liquidity could impact liquidators’ ability to profitably manage distressed collateral positions, potentially leading to bad debt accumulation. To mitigate this risk, conservative supply caps may be necessary, though this could constrain AAVE’s revenue potential.

Despite these challenges, Mantle’s position in the market is noteworthy, with TVL rankings close to major L2 solutions like Arbitrum, Base, and Optimism. This market presence, combined with the network’s improved data availability design compared to other alt-DA solutions, provides a foundation for AAVE’s expansion, albeit with careful risk management considerations.

5. Asset suggestions

5.1 Overview

We will align with @ChaosLabs on the launch asset recommendation.

5.2 Asset parameters

To be provided.

Disclaimer

This review was independently prepared by LlamaRisk, a community-led non-profit decentralized organization funded in part by the Aave DAO. LlamaRisk is not directly affiliated with the protocol(s) reviewed in this assessment and did not receive any compensation from the protocol(s) or their affiliated entities for this work.

The information provided should not be construed as legal, financial, tax, or professional advice.