# Intro to AVS Evaluation ![2024-05-03 08.59.21](https://hackmd.io/_uploads/rkUlcYGzR.jpg) *Publish Date: May 3, 2024* *We are the YieldNest Risk Team, an independent risk and research firm operated by Llama Risk. Our work serves as a public good directed toward the YieldNest DAO for the purpose to assist in AVS onboarding decisions, but may be useful for anyone interested in liquid restaking or the study of the nascent AVS ecosystem.* ## Introduction This post initiates a discussion on the emerging landscape of AVSs and how the YieldNest DAO can start to categorize and evaluate these systems both individually and as part of a portfolio of whitelisted AVS protocols. We will focus on key considerations for assessing risk and reward projections, which will be further elaborated in our forthcoming AVS Assessment Framework. Additionally, we plan to roll out a series of blog posts that will delve into specific aspects of EigenLayer and AVSs that all DAO members and YieldNest users should understand. ### What is an AVS **Actively Validated Services (AVSs)** are blockchain-based systems that use the Ethereum PoS consensus mechanism, facilitated by EigenLayer, to boost their security and efficiency. These services engage Restakers and Node Operators, registered with EigenLayer, who execute tasks essential to the AVS’s functionality. By staking tokens and supplying computing resources, Restakers and Node Operators support the integrity and robustness of the AVS. Unlike traditional Ethereum staking, where staked ETH solely maintains Ethereum's network consensus, AVSs utilize these assets and others to validate specific operations. If a Restaker or a Node Operator acts against the AVS's slashing conditions, they face penalties through asset slashing, providing a crypto-economic incentive to adhere to the AVS's rules. Conversely, compliance with these rules ensures financial rewards from the AVS they support. ### Why an AVS Any application requiring trust may consider transforming into an AVS. Using EigenLayer, developers can embed verifiable trust into their applications, turning them into AVSs. There are three forms of trust that application developers can leverage, as categorized by EigenLayer in their recent blog post, [The Three Pillars of Programmable Trust: The EigenLayer End Game](https://www.blog.eigenlayer.xyz/the-three-dimensions-of-programmable-trust/): - **Economic Trust** - Trust derived from individuals making commitments and backing their promises with financial stakes. Economic trust is trust based in capital. These types of assurances are accepted when certain commitments are backed by financial stake at-risk that makes it irrational for a rational economic actor to behave badly. - **Decentralized Trust** - Trust sourced from a network operated by independent, geographically isolated operators. Decentralized trust is trust based in decentralization. These assurances derive from having a large and distributed enough validator set that you can be comfortable that they are not all colluding. - **Ethereum Inclusion Trust** - Trust that Ethereum validators will construct and include your blocks as promised, in accordance with the consensus software they operate. While the first and second trust models absorb the economics and decentralization of the Ethereum trust network, the fact that it is the Ethereum validators who restake enables a whole suite of powerful new features where you can experiment with new opt-in features to the Ethereum protocol without modifying the core Ethereum protocol. These opt-in features open up new proposer commitments on top of the existing proposer commitments from the consensus protocol. ![image](https://hackmd.io/_uploads/SJHaSYzGC.png) Source: [blog.eigenlayer.xyz](https://www.blog.eigenlayer.xyz/the-three-dimensions-of-programmable-trust/) ### How Does an AVS Work An AVS functions by integrating with Ethereum's consensus layer via the EigenLayer protocol, enabling blockchain applications to utilize Ethereum's security features without establishing their own consensus mechanisms. This process involves: - **Restaking on EigenLayer** - AVSs leverage EigenLayer to restake Ethereum (ETH), or supported ERC20 tokens, via Restakers, directly connecting to Ethereum's security and validation mechanisms. This allows AVSs to achieve up to the same level of security as Ethereum’s native transactions, and perhaps more importantly, bootstrap security in an effective way. - **Decentralized Consensus** - Through decentralized consensus mechanism, an AVS's off-chain operations are independently validated while benefiting from Ethereum's extensive validator network. - **Operator Involvement** - Node operators secure AVS operations by restaking their ETH or supported ERC20 tokens on EigenLayer, via Restakers, receiving additional validation rewards from AVS operations on top of Ethereum rewards. This creates a financially incentivized ecosystem to support AVS networks. - **Security and Efficiency** - By connecting with Ethereum via EigenLayer, AVSs bypass the need for creating and managing separate consensus operations, significantly reducing vulnerability to attacks and operational costs. - **Network Flexibility** - AVSs are free to define their own slashing condition (i.e. what happens when an operator fails to comply with the AVS's requirements) and their own Restakers/Node Operators requirements (e.g. what assets can to be staked, how much, geographical location of operators, and more). The technical essence of AVS operation lies in its innovative use of Ethereum's established security infrastructure through EigenLayer's restaking mechanism, facilitating a secure, efficient, and scalable validation process for blockchain applications outside the EVM scope. ### AVS Product Categorization The EigenLayer ecosystem is still in its early development phase, and therefore categorization of AVS providers is exploratory and subject to revision as the industry matures. The following categories are an exercise in examining AVSs based on projected product types likely to come to market. Depending on the specific goals that an AVS is designed to achieve, we can broadly categorize them into three distinct types: **Security Enhancement**: AVS services designed to improve the security of existing blockchain architectures, such as through additional staking layers or enhanced validator commitments. **Performance Optimization**: AVSs that focus on improving the performance metrics of blockchains, such as reducing latency or increasing transaction throughput. **Functional Expansion**: Services that add new functionalities to blockchains, such as enabling new forms of consensus or introducing innovative transaction types. ### Categorization of AVS by Product Type The product classification outlined below organizes the potential solutions into seven distinct product types, each addressing specific needs within the blockchain ecosystem. **Scalability Solutions** Rollup Services are focused on enhancing blockchain scalability through off-chain transaction processing while maintaining data integrity on the main chain. Some examples: - Data Availability - Decentralized Sequencing - Fast Finality layers - MEV-aware services - Watchtowers - Based preconfirmations - Rollup deployers or RaaS (Rollup as a Service) **Interoperability Solutions** Cross-Chain Communication and Bridges enable seamless asset and information transfer between different blockchain networks. Decentralized RPC Networks facilitate communication across the blockchain network. **Applied Cryptography** Security and Privacy Enhancements utilize cryptographic techniques to secure transactions, data, and communications within the blockchain network. This encompasses methods for enhancing privacy (e.g., zero-knowledge proofs) and ensuring the security of blockchain operations, making it a comprehensive category that addresses both the protection of data and the safeguarding of transactions against unauthorized access or manipulation. Some examples: - Trusted Execution Environments (TEE) - Threshold Cryptography - Threshold Fully Homomorphic Encryption (FHE) **Computational Enhancements** Coprocessors/Computation Networks extend the computational capabilities of blockchain networks, offering specialized processing power for complex operations. Different types of coprocessors exhibit distinct cost, latency, and security characteristics. Combining different types of coprocessors can lead to an optimized user experience. Some examples: - ZK coprocessors - Optimistic coprocessors - Cryptoeconomic coprocessors ![image](https://hackmd.io/_uploads/ryjABFGfA.png) **Data and External Information Integration** Oracle Networks provide a reliable way to bring external data into the blockchain, essential for the functionality of smart contracts that depend on real-world information. **Automation Infrastructure** Keeper Networks automate essential actions within the blockchain ecosystem, such as executing smart contracts under specific conditions, to ensure efficient and reliable operations. **Development Infrastructure** AVS Tooling includes development tools and frameworks for creating, deploying, and managing AVSs, facilitating the development process and fostering innovation within the ecosystem. **Dispute Resolution** An on-demand dispute resolution system bolstered by EigenLayer validations as a business model with off-chain components that can be validated on-chain at affordable costs. Businesses (e.g., Layer 2 solutions, DEXes, Oracles, AI Companies, Cloud Compute, ZK Services) may opt-in for AVS-run arbitration. ### AVS Evaluation It can be posited that every decision in life can be analyzed through a Risk/Reward (R/R) lens. Therefore, we aim to develop a decision-making framework based on this approach. In this section, we focus on evaluating an AVS from a Restaker's perspective. This narrows the range of variables we need to consider, which should help us construct a more effective framework and make more informed decisions. The objective of this evaluation framework is not to be exhaustive, but rather to introduce the reader to a mental model that we find useful when assessing any AVS. **Risk** Engaging with an AVS involves a comprehensive array of inherent risks even before assessing the specific characteristics of the AVS. Key risks include: - Native ETH staking risks - Risks associated with EigenLayer smart contracts - Centralization risks within EigenLayer - Potential collusion among Node Operators - Risks of unintended slashing Additionally, each AVS may present its own unique risks, such as: - Vulnerabilities in AVS-specific smart contracts - Subjective conditions for slashing - Network requirements that may not be optimal - The presence of centralized components within the system - Software used by node operators that may be prone to errors **Reward** Choosing to opting into an AVS undoubtedly offers benefits, most objectively measurable through the Annual Percentage Rate (APR). If the APR is sufficiently high to offset the risks outlined previously, engaging with a specific AVS could be a prudent decision. However, it's important to recognize that this isn't merely a matter of individual preference. The entire YieldNest DAO is affected by these decisions, emphasizing the need for a collective approach to risk evaluation. To support this, we are developing a comprehensive framework that will guide the DAO in making informed, methodical decisions about AVS onboarding. This framework will provide a structured means to assess both risks and rewards effectively. ### Takeaway on AVS Evaluation We believe that the array of AVS offerings could be extensive and unpredictable. Given the rapid pace of developments, it becomes a crucial responsibility for the YieldNest DAO to identify which AVSs are appropriate for integration. This selection process involves a detailed review of each protocol's unique attributes and a strategic approach to diversifying exposure across various AVS protocols. This task will undoubtedly be ongoing and starts with a thorough understanding of the risks and benefits associated with each AVS, enabling the DAO to make well-informed onboarding decisions.