A quick look at the 18 projects that won the Ethereum Foundation’s “Account Abstraction” funding round

Author: ERC-4337; Compiler: Deep Tide TechFlow

The ERC-4337 Account Abstraction Funding Project has been successfully concluded. The grant program will support 18 teams building diverse projects around ERC-4337 (also known as account abstraction).

Each team’s project is uniquely aligned with the goals of the ERC-4337 AA grant, and we believe the results of these efforts will ripple throughout the ecosystem, inspiring new ideas and collaboration opportunities.

The passionate and talented individuals driving these projects are at the heart of this funding round’s success, demonstrating a deep understanding of the Ethereum ethos and a commitment to pushing the boundaries of possibility. Recipients of this round of funding have a variety of skills and experiences and are collectively committed to advancing Ethereum’s capabilities.

To get a feel for individual projects and leadership teams, we wrote project overviews provided by the teams themselves (in no particular order). As these projects move forward, we eagerly await their progress and potential impact on Ethereum’s trajectory.

ZeroDev Kernel

ZeroDev Kernel, or Kernel for short, is a smart contract account designed from scratch with the following features:

• For AA wallet - Kernel is compatible with ERC-4337;
• Efficient - We pay great attention to optimizing its Gas usage;
• Modularity - Kernel can be extended through “plugins”.

Plug-ins are smart contracts that customize account behavior. We recognize that it is impossible to predict users’ needs for smart contract accounts, because in the process of using the accounts, they will interact with many different DApps. Therefore, the only solution is to make the account scalable so that users can add new features to their account over time.

As of this writing, Kernel is the most widely used AA account on all major EVM networks (Ethereum/Polygon/Arbitrum/Optimism/Base). Kernel is a project of ZeroDev that provides infrastructure and tools for developers to integrate AA.

Etherspot 的 Skandha Bundler

ERC-4337 introduces a separate memory pool dedicated to activities performed by account-abstracted wallets. In addition, it proposes a novel concept called “bundler” that merges user operations into the traditional trading framework. This bundler-driven approach enables account abstraction to be performed and implemented without changing the underlying protocol. This mechanism also lays the foundation for seamless integration in the future. In this case, the importance of the P2P interface in the bundler becomes obvious. By integrating the P2P interface, user actions received by a specific bundler will be propagated among all interconnected nodes. Therefore, even if a specific bundler exhibits malfunctioning behavior or transaction review, these user actions can still be processed by other bundlers in the network.

The Etherspot team has helped the Ethereum Foundation develop a shared memory pool protocol specification and is using Etherspot’s bundler implementation to pilot a shared memory pool (P2P interface). Etherspot’s bundler is named Skandha. It is already available on six testnets and eleven mainnets, including Ethereum, Polygon, Optimism, Arbitrum, Fuse, Mantle, Gnosis, Base, Avalanche, BNB, and Linea.

MynaWallet

MynaWallet is a contract account integrated with the Japanese government-issued identification card (Myna Card), which is currently held by approximately**100 million residents. **MynaWallet enables users to seamlessly make token payments, transfer or receive NFTs, and perform any transaction supported by their account, without ever needing to be aware of the existence of cryptocurrencies.

This has the potential to make cryptocurrencies accessible to most Japanese residents. We would be delighted if this initiative could set a precedent that inspires other countries to conduct similar experiments, connect their national identity infrastructure with Ethereum, and introduce their residents to the cryptocurrency space.

Blocto: AA support for web3.js

We believe our Web3.js project not only aligns with the goals of ERC-4337 AA funding, but plays an important role in the broader field of dApp development.

In order to promote the development of ERC-4337, we decided to share 4337 specific RPC methods through Web3.js, such as sendUserOperation, estimateUserOperationGas, getUserOperationByHash, getUserOperationReceipt and supportedEntryPoints.

By doing this, developers can use these methods to focus on what they really care about: amazing dApps. For example, if dApps want to provide a better user experience, they can change the calling method from sendTransaction to sendUserOperation as long as the wallet provider supports this feature.

Blocto is a firm believer in simplifying these processes, facilitating the growth of a supporting dApp ecosystem, and ultimately enabling a new era of decentralized applications that is as much developer-focused as it is user-focused.

Silius

Silius is a modular and efficient Rust implementation of the ERC-4337 bundler. The project is fully aligned with the broader ERC vision and ecosystem, keeping pace with specification changes and supporting upcoming features such as P2P protocols and alternative mempools. The bundler is structured so that its core building blocks - the main components and libraries - can be reused as LEGO bricks in other AA projects in the ecosystem. Silius also contributes to the Rust Ethereum ecosystem by building on and extending on other Rust libraries and projects.

Schnorrkel.js

We should emphasize the fact that Schnorr signatures are enabled through the account abstraction. Without account abstraction, each dapp would have to write Schnorr signature handling in its contract, and explain to the world and EOA wallets why they had to sign two messages (ECDSA for the mempool, Schnorr for the actual transaction), which makes using It looks very bulky. All these problems are solved by account abstraction. Just sign the UserOperation with Schnorr and pass it to the payment principal. Schnorr signature verification is handled by the AA smart contract. In other words, AA enables different signature schemes.

Why use multi-signature Schnorr?

• Complexity is handled off-chain, making on-chain Schnorr verification very cheap (approximately 3k gas);
• Privacy - On-chain, no one knows the number of participants since they are represented by a single address.

QR-Seal is a Schnorr multi-signature wallet that is fully compatible with ERC-4337. It uses the Schnorr library that won the grant, and the AA and payment subjects use the Ambire Wallet contract and Pimlico. We participated in the ETH Prague Hackathon and won an award in the ETH Future category with our QR seal. This hackathon highlights how two people created a multi-signature wallet and passed data via a QR code.

Ambire: Self-hosted email/password authentication via DKIM

Email signup is a viable method of introducing new users to the crypto world as it provides a familiar process for access and recovery using email. We will provide an easier and more native way to add recovery mechanisms or transaction signatures to account abstraction wallets, while providing greater security. This type of recovery or transaction signing will enhance self-custody and user experience for new users in Web3.

Such a library will reduce the effort required by other developers to provide their users with a more user-friendly application while they focus on AA solutions.

Furthermore, we believe that such applications will greatly attract new users to the Web3 world.

The library will be fully open source and licensed under a permissive license like MIT, like all our current efforts on DKIM authentication or email/password accounts.

Ambire is a web-based account abstraction wallet that launched in late 2021 and is the first wallet to offer self-hosted account creation via email/password authentication, thanks to the underlying 2/2 multisig.

After a year of running on mainnet and other EVM networks, we think email/password accounts demonstrate how account abstraction can make Ethereum more inclusive. We’ve found that email-based registration works well for those who don’t have the technical skills or patience to deal with seed phrases or browser extensions.

Ambire Smart Accounts provide the option of multiple signers. We support two types of signers. The first is email/password signers and the second is EOA such as hardware wallets, browser extensions (Web3 wallets), etc.

When using an email/password signer, Ambire creates a 2/2 multisig behind the scenes for you, where one key lives on your device and is encrypted with a passphrase, while the other key verifies the email confirmation code and stored on the backend HSM. Two keys are required to control the account, but only one key is needed to trigger the scheduled recovery process.

We provide recovery options for email/password accounts in case passwords or email access are lost. If only one key signs a transaction, the contract will trigger a 72-hour timelock recovery. After 72 hours, the missing key in the transaction will be replaced with a new key provided by the user in the same transaction.

IoTeX: Account abstraction wallet based on zero-knowledge proof

Utilizing zero-knowledge proofs (ZKP), our account abstraction wallet based on zero-knowledge proofs enhances and broadens the potential of on-chain games, DeFi, DeSocial, DePIN and other applications, establishing interoperability between Ethereum and real-world applications.

The wallet uses zkSNARKs to confirm account access: if the user has the correct password, they can use the prover to generate a valid proof to unlock the account. The on-chain smart contract then verifies this proof to authenticate the transaction.

Our forward vision involves applying zero-knowledge proofs to multiple aspects of the account abstraction wallet, including wallet operations via web2 authentication, social recovery using web2 social media state, and initializing zero-balance wallets using pervasive web2 payment channels.

The project team consists of two senior researchers from the IoTeX team.

SixdegreeLab: On-chain data analysis

The original vision of the project was to provide Web3 data analysts with the necessary knowledge and skills to adapt to the changes in the data structure of Ethereum and EVM chains caused by ERC-4337, and to maintain effective on-chain data analysis. Specifically, the project will:

1. Provide several data indicator dashboards to help account abstraction practitioners understand the ecological data situation. At the same time, create an aggregated data table to assist the analyst’s work.
2. Provide an educational tutorial document covering the impact of account abstraction on on-chain data analysis, which will be open to Web3 data analysts to enable them to adapt to new data structures and effectively conduct on-chain data analysis in Web3 .
3. As more and more players get involved in account abstraction, including infrastructure, dApps, and mainstream users, we need a data perspective report to illustrate the current ERC-4337 adoption situation.

By addressing the impact of ERC-4337 on on-chain data analysis, and exploring native account abstraction mechanisms in other Layer2 and EVM chains, this project can provide valuable insights and resources for data analysts working in the Ethereum ecosystem. This in turn allows for more effective identification and mitigation of vulnerabilities and performance issues on the Ethereum platform, helping to ensure its continued success and growth.

Overall, this project can help ensure that on-chain data analysis remains a valuable and effective tool for the Ethereum community, and by providing insights and resources for data analysts working in Ethereum, Layer2 and EVM chains, Contribute to the growth and success of the Ethereum ecosystem.

Jam: Picnic’s DeFi transaction batch processor

Jam is an open-source API for generating DeFi operations for ERC-4337 applications. The goal is to enable developers to easily offer a variety of DeFi products through a simple API, while making DeFi more accessible to end users. You can ask Jam how to perform a specific DeFi operation and receive calldata that performs all the required steps to complete in a single transaction.

We (Picnic) have been working with various DeFi integrations for the past two years and felt the difficulty of maintaining such infrastructure. We believe the best way forward is to turn it into a public good, and funding from the Ethereum Foundation is an excellent way to coordinate with the community.

TokenSight: Account abstraction multi-factor authentication using WebAuthn

The problem TokenSight is trying to solve is related to on-chain smart contract wallet security, and the approach is mainly focused on two overlapping directions - creating an ERC-4337 smart contract wallet with a WebAuthn device and providing development for existing ERC-4337 wallets. author tool to add the WebAuthn facility as a multi-factor transaction authorization mechanism.

Our approach is based on a fully on-chain implementation of the Gas cost-optimized WebAuthn signature verification scheme without using any MPC architecture, thereby providing maximum security and decentralization.

The team for this project consists of Nasi and Blagoj, who have many years of experience in the blockchain industry. We are passionate about smart contract security and we believe that providing maximum security and a seamless user experience for users’ funds in their smart contract wallets is important to attracting the next billion users. We are currently building Tokensight, which will be the first application to use our open source WebAuthn wallet.

ZK-Team: PriFi Labs’ privacy-preserving account abstraction for teams

A valuable aspect of account abstraction is its ability to facilitate shared account ownership among team members, enabling them to transact seamlessly as a unified entity. Organizations can create an abstract account and establish predefined rules to manage authorized individuals and their assigned spending limits. In a simple setup, team members’ addresses and their corresponding allowances might be written to the blockchain. Doing so raises a legitimate privacy concern, as organizations may be reluctant to publicly disclose this sensitive information.

ZK-Team is a proof-of-concept of a privacy-preserving abstract account that allows organizations to manage team members while maintaining individual privacy. By utilizing zero-knowledge proofs, ZK-Team enables transactions that ensure the secrecy of team members’ unique addresses and their associated perks. We expect our project to be a one-stop solution for organizations managing teams and assets, as well as a reference for Ethereum developers who want to use zero-knowledge proofs to protect the privacy of information stored in the account abstraction.

PriFi Labs is a Toronto, Canada-based startup researching how to enhance the security and privacy of blockchain applications.

Rhinestone: Modular Account Abstraction

Account abstraction promises to provide enhanced security and meaningful user experience improvements without compromising the Web3 philosophy. At Rhinestone, we believe the future of account abstraction is modular, where modules are smart contracts that extend smart account functionality, providing a new platform for permissionless wallet innovation.

However, opening smart accounts to third-party developers through modules increases the attack surface of user accounts. We are developing a standard for registering modules and declarations to make statements about the security of those modules. This registry serves as a trust delegation mechanism for smart accounts to query before installing third-party modules to increase user security.

AAA: Account Abstraction Afterhours

At AAA, it all started with a simple idea - to raise awareness about AA and 4337. While we may no longer be coding in person, our passion for technology and commitment to mass crypto adoption remains strong. We firmly believe that account abstraction is key to driving widespread adoption.

What started as a developer meetup has grown into a series of impactful events hosted on three continents. These activities provide a platform for knowledge sharing and community building, perfectly aligned with the goals of the ERC-4337 AA grant.

Looking ahead, we are excited to further expand our reach. Educational videos and interactive tutorials are coming soon, allowing us to empower more people to engage with and understand crypto. We are honored to be a part of this journey and committed to making meaningful contributions to the field of encryption technology.

Unpacking 4337 - A series of technical articles

The authors of this series of articles received a grant to help understand how components of EIP-4337 operate. The goal is to reduce the time and effort required for developers and researchers to catch up on progress and start focusing on specific areas they are interested in contributing further.

The three themes are:

1. User operation life cycle: the journey of the user operation in the packager and the different functions called by the packager in the entry point to verify and finally submit it to the chain.
2. Understand the different Gas parameters: There are multiple Gas parameters specified at the packer, memory pool, and user action levels (listed below). Explain these parameters and their impact.
3. Explain the different calldata parameters: input transaction calldata, UserOp.calldata, multi-send Txns calldata.

Hexlink: EIP-4972 - Link ENS to ERC-4337 accounts

Hexlink aims to provide all ENS users with an AA-compatible smart account that can be used directly through EIP-4972. Through Hexlink, each ENS name can derive a unique AA-compatible smart account that can be deployed only by the domain owner and can be used to receive cryptocurrency instantly. Considering that most ENS names are currently controlled by EOA, we believe this dual-account setup will facilitate the migration from EOA to smart accounts, converting all existing ENS users into AA users, thereby promoting wider adoption of AA.

UniPass: OpenID authentication module

Account abstraction enables developers to customize verification logic and leverage various Internet infrastructure for user authentication. Our UniPass team is introducing a novel user authentication solution by verifying on-chain signatures of ID tokens for the OpenID Connect protocol - an OpenID authentication solution for new user onboarding and recovery.

This solution works in conjunction with ERC-4337 to provide the convenience of one-click OAuth login and account creation to new users using social logins such as Google and Apple. Additionally, it simplifies the recovery process in the event a key is lost or compromised. This project aims to build a library that enables ERC-4337 wallet developers to easily integrate OpenID authentication into their products.

Clave: Promote secp256r1 curve support for AA

Although account abstraction solutions are emerging, providing a better user experience without compromising security remains one of the major challenges. The Clave team (formerly Opclave) is researching and developing tools that leverage mobile device native software and hardware solutions as signers, elevating mobile device wallets to the security level of hardware wallets by maximizing the user experience. The following are some of the objectives of this scope of authority:

• Share research results utilizing solutions such as Secure Enclave, WebAuthn, Passkeys and more.
• Provides sample code for using these solutions as transaction signers.
• Provides smart contract sample code for processing these signed transactions.
• Provides sample account contracts managed by these solutions by integrating Solidity validators and precompiled contracts.
• Promoted EIP-7212 to improve these solutions for efficient use in EVM.

The Clave team includes developers and researchers who first introduced the concept at the “ETHGlobal Scaling Ethereum 2023” hackathon. Now, the team is working on leveraging a mobile device’s trusted execution environment as a signer to provide accounts driven by the account abstraction.