Contents

1. Before AA
2. AA Transaction Flow
3. Components of AA
4. AA Stack
5. What will AA bring?
6. Investment Thesis, where will value accrue?
7. Other
8. Closing Remarks

Written by Calvin and Thomas

In the ever-evolving realm of Web3, a momentous concept is stirring noteworthy buzz within the crypto community, poised to unleash a tidal wave of new adopters. This ‘transformative’ innovation is none other than Account Abstraction — a powerful catalyst, redefining the very fabric of our interaction with Ethereum.

The conventional days of Externally Owned Accounts (EOAs), exemplified by popular wallets like MetaMask, wherein users bear the burden of safeguarding their private seed phrases may be gone. Embracing the tenets of progress and innovation within the crypto domain, Account Abstraction offers an alternative for seamless Ethereum engagement.

Account Abstraction (AA) allows users to embark on a journey where user-centric experience assumes paramount significance, and impregnable security becomes the cornerstone. AA seeks to dismantle the limitations of antiquated practices while fortifying the sanctity of assets for seasoned enthusiasts and risk-averse novice users alike.

TLDR

• Introduction:
  Account Abstraction (AA) stands at the forefront of Ethereum’s evolution, promising to redefine how users interact with the platform. By offering an enhanced and more user-friendly alternative to the traditional Externally Owned Accounts (EOAs) like MetaMask, AA is poised to significantly elevate the Web3 user experience.
• Ethereum’s Current Account Framework:
  Ethereum’s ecosystem primarily revolves around two account types. The Externally Owned Accounts (EOAs) are controlled by individual private keys, allowing users direct interactions. On the other hand, Contract Accounts (CAs) are driven by smart contract codes, enabling more complex operations and interactions.
• The AA Stack and Its Significance:
  The emergence of projects under the ERC 4337 banner is reshaping the blockchain landscape. These projects encompass a wide range of blockchain interactions, from facilitating Bridges and Ramps for seamless asset transfers to introducing Gas Abstraction mechanisms that simplify transaction costs. The AA stack aims to streamline and enhance these interactions, making them more intuitive for end-users.
• Pioneering Features Introduced by AA:

1. Trusted Sessions: Drawing parallels with online shopping experiences, AA introduces the concept of ‘Trusted Sessions.’ Users can set predefined rules for their dApp interactions, eliminating the need to sign every transaction.
2. Scheduled Transfers: AA paves the way for innovative financial solutions, such as self-custodial scheduled payments. This feature holds the potential to revolutionize on-chain subscription models, offering unprecedented flexibility.
3. Multi-Party Approval with MPC: By integrating Multi-Party Computation (MPC) wallets into the AA framework, fund access becomes more secure and streamlined. External services can now withdraw funds without the cumbersome process of gathering multiple keyholder signatures.

• Potential Risks and Challenges:
  While AA offers numerous advantages, it’s not without its challenges. Concerns have been raised regarding the traceability of funds, potential new attack vectors that malicious actors might exploit, and the increased costs associated with AA’s implementation. Additionally, there’s a looming fear of infrastructure centralization, which could undermine the decentralized ethos of blockchain
• Investment Insights from OP Crypto:
  OP Crypto, with its keen market acumen, is closely monitoring developments in the Account Layer. They’ve shown their support for Particle Network, a project dedicated to refining AA’s modularity. The emphasis is clear: for AA to gain widespread adoption, solutions that demystify its underlying complexities are paramount.
• The Promise of ERC-6900:
  Building on the foundations laid by ERC-4337, ERC-6900 was introduced in April 2023. This new standard aims to bring about greater standardization in account utilities, ensuring that AA’s benefits are more accessible and modular for the broader community.
• Final Thoughts:
  The AA landscape is vibrant and dynamic, with a plethora of talented builders continuously pushing its boundaries. Despite the inherent market challenges, there’s a tangible sense of optimism surrounding AA’s potential, especially given the innovations brought about by ERC-4337. With active investments and a keen interest in fostering collaborations, OP Crypto is firmly positioned at the heart of this evolution.

1. Before AA

Two account types exist on Ethereum today:

1. Externally Owned Accounts (EOAs)

2. Contract Accounts (CAs)

The below chart outlines some of the primary differences between EOAs and CAs:

As mentioned, EOAs are accounts generated using wallet software such as MetaMask or Rainbow, which are managed by their respective public and private keys. Such accounts have relatively limited capabilities as their use typically does not exceed transaction initiation and gas fee payment.

2. AA Transaction Flow

Below is a high-level diagram presenting the ERC-4337 transaction flow:

1. A user requests to complete a transaction via a supported dApp
2. Wallet Software creates a UserOperation

a. UserOperation is displayed to the user

b. User is prompted to sign

c. UserOperation is sent to a Bundler

1. The Bundler receives the UserOperation

a. The Bundler verifies UserOperation: i. via the EntryPoint

b. UserOperation is added to the alternative mempool

c. Block builder adds as a block to the blockchain

Below is an in-depth diagram presenting the ERC-4337 transaction flow:

Image from Visa’s “Rethink Digital Transactions with Account Abstraction”

Some of the above terms may be unfamiliar to even crypto-native users, as they are concepts native to Account Abstraction, introduced via ERC-4337. Now we will discuss these new terms and concepts which separate AA from the current EOA standard.

3. Components of AA

UserOperation (UserOp) is a pseudo-transaction object that represents a user’s desired transaction intent. UserOperations are technically NOT transactions, although may be referred to as an “ERC-4337 transaction.”

§ UserOperation Mempool: A mempool specifically for UserOperations. Consists of a permissionless peer-to-peer network where nodes use logic to determine whether the UserOp will be sent to peers. This is a new stop (compared to traditional mempool) for a UserOp before it lands on-chain.

§ UserOps that only use whitelisted Paymasters and/or Aggregators are placed into an alternate mempool, where Bundlers that support the whitelisted Paymasters/Aggregators may use UserOps from alternate mempool. UserOps from main and alternate mempools can be bundled together.

Bundler

Bundlers bundle multiple UserOps together and sends said bundle for execution via the EntryPoint. This event looks like an EOA to smart contract transaction, but in this case the “from” address is the Bundler, and the “to” address is the EntryPoint contract.

UserOps are chosen by Bundlers based on fee-prioritization; the same logic that Ethereum block builders use today. Since Bundlers are the sender, they are responsible for paying the gas fee, but are compensated through individual UserOp execution fees.

Bundlers are the nodes that submit a UserOperation on-chain. There are both public and private bundlers, each serving the same function, yet bundling UserOps from different sources. Public bundlers receive UserOperations from all unverified sources in a decentralized manner, which requires them to or undergo a filtering process in order to reject transactions with harmful potential. In particular, they are crucial in preventing Denial-of-Service (DoS) attacks through the management of paymasters’ reputation. The functions of public bundlers all have a cost, in either computation, gas, or storage, to perform properly.

Conversely, private bundlers are bundlers built to only bundle the UserOperations from a specific source. For example, a wallet provider may construct private bundlers that only accept UserOperations from users of that wallet. This method allows some processes to be bypassed resulting in lower usage costs to users through saved gas fees. Additionally, private bundles can provide speedier mempool inclusion as well as the potential for MEV resistance transactions.

Paymaster

A paymaster is responsible for the sponsorship of gas fees for EVM transactions via Account Abstraction through the EntryPoint contract. Paymasters also permit payment of gas fees in ERC-20, credit card, subscription, etc. so users need not sacrifice custody or hold Eth. The logic of gas abstraction is defined by the Paymaster, and are deployed by dApps that want to sponsor their user’s gas fees.

DApp developers may wish to use a paymaster because sponsoring gas fees lowers the barrier to participate, making it easier to acquire more users. In order for developers to incorporate a sponsorship paymaster, they must set the policies regarding which UserOps can be sponsored. The number of rules supported, however, is entirely up to the paymaster provider.

Allowing gas payment in ERC-20 serves a similar purpose: improving the user experience. The ability to pay in stablecoins is more appealing to a user as opposed to more volatile assets like Eth. Having to hold Eth in a wallet to pay gas fees is not conducive to new user growth as the process of purchasing Eth may be enough to prevent non-crypto-native users from continuing further.

Note: Paymaster may also be custom built and operated, but below we list some of the leading paymaster providers.

Notable Projects:

StackUp’s Plug-n-Play Paymaster

Alchemy’s Gas Manager

Pimlico’s Verifying Paymaster

EntryPoint

The EntryPoint verifies and executes the UserOperations sent to it by the Bundler. They serve as the entry point that all ERC-4337 compliant smart contract wallets use to transact on the EVM. This can be most aptly compared to a single staking deposit contract. Logic used by smart contract wallets is simplified when using an EntryPoint, abstracting away the complicated smart contract functions to the EntryPoint instead of the wallet. Essentially, it is the EntryPoint mechanism that allows wallets to function as smart contracts.

“…Give[s] wallet infrastructure providers more options for providing smart wallet features like account recovery, native multi-sigs, and covering gas fees for users.”

• Michael Lewellen, Head of Solutions Architecture @ OpenZeppelin

Wallets

Wallets are at the forefront of the Account Abstraction discussion as they have a direct connection with the user, and are straightforwardly involved in offering Web3 users a safer and more enjoyable experience. Smart Contract Wallets allow users to customize the way in which they manage their assets, as made possible through Account Abstraction.

Before ERC-4337, third-party relayers were required to abstract away the details of how and when an EOA functions, or with AA-native Layer-2s such as StarkNet or ZKSync.

Wallet Operation Comparison

4. AA Stack

We can broadly understand the projects building in the ERC 4337 realm in the follow buckets. These layers work in conjunction to cover majority of end user interactions with blockchain.

3.1 User Interface (UI) Layer

The IU Layer primarily offers the toolkits necessary to leverage the improved UX experience that AA can bring to web3. This is the layer in which users are onboarded via email address, phone number, wallet address, or social media profile. Additionally, projects building in this layer provide the tools for app developers to allow users to create their own wallet (“embedded wallet”) into their app. These wallets can be customized to match dApp branding and provide additional features.

In order to be considered a “wallet,” there must be a semblance of UI. Developers connect and/or create interfaces and applications that allow users to interact with blockchain accounts.

Embedded Smart Wallet Creation:

“Once a user logs in with email (One Time Password sent to ensure it is a functional and active email address prevents errors in customer communication and ensures that customer data is accurate) an embedded wallet is created for them, a key pair is generated for that wallet. The key pair is then used to generate a smart contract wallet that uses the private key as the owner. In essence, you use the private key from the embedded wallet to control a more sophisticated smart wallet.

The resulting wallet is returned as a provider, and can be interacted with in a similar way to an EOA, with the flexibility to define things such as gas sponsorship as part of the wallet initiation”

• Dynamic.xyz.

More specifically, the Wallet (Account) contains three different components, each with their own specific function.

1. Wallet Contract: The contract that represents a user’s account

2. Wallet Deployer: The contract that creates a wallet contract

3. Wallet Software: The UX that sends UserOperations to the nodes (Bundler)

Notable projects:

Particle Network, Privy, Dynamic, Crossmint

3.2 Account Layer

The Account Layer can be divided into two sublayers: AA Infrastructure, and Smart Account.

1. AA Infrastructure

Infrastructure in AA refers to the underlying components and mechanisms which allow for Account Abstraction to occur. The primary function of this layer is to supply Bundler, Paymaster, and EntryPoint services that seamlessly interact with AA wallet providers. Earlier, we outlined the how these components fit into the wider transaction flow of AA.

Notable projects:

Alto Bundler (Pimlico)

Alchemy Rundler

Infinitism Bundler

Voltaire (Candide)

StackUp Bundler

AA-Bundler

Skandha (Etherspot)

1. Smart Account

• Wallet specific to a type of smart contract that implements AA. As long as the logic is able to be realized by the Ethereum smart contract, the account can provide the desired AA functions via customization, plug-ins, etc. This sub-layer can be aptly described at the backend of an AA wallet, which effectively sends UserOperations to the services provided by AA Infra providers.
• The “Smart Account” is what allows wallets to be controlled via programmable smart contracts as opposed to private keys. This added composability allows for customizable features such as performing a certain task, supporting various authentication methods, etc.

Notable projects:

Particle Network

Biconomy

Sequence

Clave

ZeroDev

Safe{Core}

3.3 Key Layer

The Key Layer contains pluggable authentication infrastructure for Web3 applications and wallets.

Alternative forms of log-in and wallet creation methods are a focal point of Account Abstraction wallets. Social and email login are two of the highlighted forms of login expected to grow the Web3 user base in the coming years as these methods can simplify the onboarding process with Web2-esque features to ease their transition into Web3.

When a user logs in to a dApp using a social media account or email, authentication providers like Particle and Web3Auth can create a wallet for said user without a private key. The aforementioned providers guarantee this security through Multi-Party Computing (MPC) based signatures.

The SDKs for these features are available for use from authentication providers for integration in their dApps. A necessary step for implementing Account Abstraction into a dApp is social login. let’s discuss the relationship between Web3Auth’s SDK and an integrating application.

When a user triggers a login from one of the various login providers (through an application with Web3Auth integrated), an ID token is submitted to the Web3Auth SDK. That ID token is then passed to the Auth Network for validation and in return, a user’s private key is split up into multiple shards where the user needs only 2 of the total key shares to recover their private key.

Notable Projects:

Particle Network

Web3Auth

Portal

0xPass

5. What will AA bring?

Better UX

Interacting with applications in crypto is not necessarily a simple feat. If a user wishes to use multiple applications, there can be as many as five separate steps involved to complete the desired function. More importantly, relying on a 12-word seed phrase to maintain control of one’s funds is unreasonable bestow on people wishing to enter the space when the already risky nature of crypto is already a daunting barrier. Account Abstraction permits the recovery of funds via social media applications, email, friends and family, etc. In other words, losing the private key is not the end-all.

Lower Fees

Account Abstractions allows users to pay fees in any contractually supported token or even pay no fees at all. As outlined in the above chart, with an EOA, users must maintain an ETH balance to complete a transaction as a gas fee must be paid by the user (initiator). With AA, the incorporation of Paymasters with ERC-4337, allows gas fees to be sponsored/subsidized, so users need not maintain an ETH balance within their wallet. Also, having the ability to pay with various types of tokens allows for greater freedom and ease of use when operating within the crypto space.

Batching Complex Actions (aka Multicall)

Instead of constantly pressing “approve” and “confirm” buttons whenever making a transaction, Account Abstraction can be programmed to batch multiple operations to be signed at once. This can make transactions both cheaper and faster.

Trusted Sessions

Account Abstraction allows users to pre-prove rules for interaction with a dApp so they do not need to sign every specific transaction and will open up better UX possibilities for end users. This feature can be analogized to online shopping, where users place multiple items in their basket, then transact all at once.

Scheduled Transfers

Through Account Abstraction, self-custodial scheduled payments and recurring transfers are made possible. This ability presents multiple potential use cases spanning from salary payments to on-chain subscription models.

Multi-Party Approval

MPC (Multi-Party Computation) wallets can work alongside AA to give users a more secure key management system. MPC allows for the decomposition of a wallet’s keypairs, which are then distributed to select trusted personnel. AA in tandem with MPC for example, allows an external service like a vendor or hosting tool to automatically withdraw funds from the account without needing the majority of keyholders’ signature.

6. Investment Thesis, where will value accrue?

While still in its infancy, determining where the most value will accrue within the AA stack is difficult, as technically there is no defined standard of which the industry submits to. However, as the stack stands today, it seems most probable that AA infrastructure providers have the most to gain when AA enters the mainstream.

There are currently a few active AA infrastructure providers that represent the easiest form of AA implementation for new and existing dApps. Bundler and Paymaster providers have the highest potential financial upside as these services are required to implement Account Abstraction, and therefore have a moat-by-default. To the community’s dismay, there exists the opportunity for these providers to form an oligopoly of sorts, whereby dApps within to adopt AA, must outsource these services, instead of building their own private bundlers and paymasters, which costly and complex.

In theory, anyone can operate a bundler, but operating a bundler involves technological complexity that developers of new projects likely won’t have the time or funds to take on, as they can easily utilize any of the currently available services on the market. While oligopolies go against the ethos of Web3 as a whole, this problem will need to be solved to prevent the encouragement of EVM centralization.

With an early-stage, open-source development like ERC-4337, the parties making advancements possible are the most probable area of value accrual.

An additional potential area of value accrual: AA Wallet SDK providers.

Our thesis at OP Crypto has not been formalized, but we keeping tabs on Account Layer developments. We hope to see projects begin tackling the current issues within the infrastructure layer, however since the concept is so new, such developments will take significantly more time.

We are a proud backer of Particle Network, who have shown incredible dedication to optimizing the modularity of Account Abstraction. They are experts in the space and are building cutting edge products to increase accessibility and efficiency of Web3 and increase usability for veterans and novices.

Solutions that have the ability to abstract away the various underlying complexities are still needed. The experience for developers to integrate AA into their dApps remains more complex than it should be to encourage widespread adoption. Particle, among other such projects are working intently on this matter, and we are extremely excited to see what solutions are built to overcome integration hurdles.

7. Other

Modular Account Abstraction with ERC-6900

The standardization of smart accounts in important to further developing their utility. The newly introduced ERC-6900 (April 2023) builds upon the features introduced by ERC-4337, and seeks to offer standardization. ERC-6900 serves to extend the functionalities of ERC-4337 by increasing the modularity of an account’s utilities. Increases in flexibility and adaptability can be attributed to its ability to support implementations through plugins.

Risks

While we have discussed the intricacies of Account Abstraction and now have a sense of how it represents progress in the realm of user experience and functionality, there are concerns about this new framework and support has certainly not been unanimous. While the below list is not exhaustive, it contains some of the primary risks visible today.

• Trace of Funds
• New Attack Vectors
• More Expensive to use
• Lack of standardization
• Centralization of infrastructure providers
• Social recovery and 2FA can be compromised and may require use of centralized counterparties

8. Closing Remarks

The Account Abstraction landscape is rapidly evolving, and will continue to evolve for the foreseeable future. So many talented builders are working on new applications and use-cases for the new technology that we are sure to see tangible evidence of its benefits in the short term. While the current market has made scaling blockchain-based projects more difficult, we are optimistic about what Account Abstraction, and more specifically ERC-4337 will do to mitigate some of the many pitfalls of the current Web3 environment.

We are actively investing in AA and AA-related projects and encourage all those building in the space to reach out to us! We are always willing to assist builders however we can, whenever we can!

Disclaimer：

1. This article is reprinted from [Inception Capital], All copyrights belong to the original author [ Calvin and Thomas]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned, copying, distributing, or plagiarizing the translated articles is prohibited.