Introduction
In the context of EigenLayer, an autonomous verifiable service (AVS) refers to a decentralized service that utilizes the security and validation power of the Ethereum staking mechanism through restaking. These services consist of on-chain verification contracts and an off-chain operator network that executes tasks initiated via on-chain contracts, direct operator communication, or task aggregator entities.
The proliferation of AVSs presents node operators with the challenge of selecting the most suitable services for their staked ETH. This AVS Evaluation Framework provides a structured methodology to facilitate this decision-making process by systematically assessing services based on technical, economic, and strategic factors.
Framework Objective and Rationale
The primary objective of this evaluation framework is to provide node operators with a structured approach for assessing AVSs, identifying suitable candidates for restaking while prioritizing risk minimization and high rewards. This framework is predicated on the principle that excessive risk-taking can result in substantial financial losses.
The framework is designed to mitigate risks associated with long-term commitments to AVSs, particularly addressing potential consequences of security incidents or bugs that may impede service exit. By prioritizing AVSs that offer high flexibility while providing high rewards and infrastructure stability, operators can minimize stake loss risk and maintain adaptability to changing circumstances.
Selection Workflow
The evaluation framework follows a structured workflow:
- Framework Calibration: Users assign weights to criteria based on their risk tolerance. These weights are globally applicable across all AVSs but can be revised over time to adapt to evolving requirements.
- AVS Scoring: Each AVS project receives a score (0-100) for each criterion, serving as a metric of excellence, with 100 signifying optimal opportunity for the node operator evaluating a given AVS. These scores are project-specific and subject to revision.
- Project Selection: Following calibration and scoring, aggregate scores are computed for each project, creating a comprehensive ranking. Users establish a threshold value or "cut-line" distinguishing projects suitable for re-staking.
- Restake Distribution: Operators allocate restaking resources across selected projects to optimize asset deployment.
- Active Monitoring: The framework functions as a dynamic process rather than a one-time evaluation. Various events may necessitate returning to initial stages for re-evaluation, including protocol-wide changes or AVS-specific events.
Weighting and Scoring Methodology
The evaluation utilizes a weighted scoring system across three categories: Technical, Economic, and Strategic. The aggregate score Pj for each AVS project j is calculated using the following formula:
Pj = ∑(Wi * Sij)
Where:
- Wi represents the weight assigned to criterion i
- Sij represents the score (0-100) assigned to criterion i for project j
- ∑ indicates summation across all criteria
This approach normalizes scores according to assigned importance and produces a final score reflecting the AVS project's suitability based on the operator's priorities.
Evaluation Categories
Technical Criteria
Technical aspects provide the means to mitigate risks and encompass five key areas:
Imposed Commitment
This metric assesses whether an AVS requires commitments beyond the standard 14-day deallocation delay inherent to EigenLayer. This includes direct commitments such as extended lockup periods and indirect commitments like specialized hardware requirements. For instance, a given AVS could demand GPU infrastructure, where the most economical cloud GPU nodes require extended rental commitments ranging from one month to a year, significantly prolonging transition periods between services.
Slashing Conditions
Evaluates the severity and clarity of conditions under which operators may lose staked assets. The framework examines these conditions relative to Ethereum's base layer standards, seeking balance between system integrity and proportionality. Clear documentation of slashing triggers, preventative mechanisms, and historical incident data contribute to this assessment.
Security and Audits
Examines the robustness of security measures, including the number and quality of security audits, reputation of auditing firms, open-source code availability, bug bounty programs, and security incident history.
Infrastructure Requirements
Assesses the physical hardware demands, including computational requirements, specialized hardware needs, bandwidth specifications, and scaling requirements based on stake amounts. EigenDA serves as an example where storage and network bandwidth requirements adjust dynamically according to an operator's restake amount.
Software Requirements
Evaluates setup complexity, documentation quality, deployment automation, and troubleshooting capabilities. This includes assessment of Docker containerization support, deployment scripts, and the time required to achieve operational readiness.
Economic Criteria
Economic factors relate directly to reward potential and stability, encompassing six criteria:
Total Value Locked (TVL)
Serves as an indicator of community interest and economic stability. Higher TVL typically indicates a more active community and stable economy, while smaller TVL may suggest volatility. The framework considers absolute TVL values, growth rates, relative positioning within the ecosystem, and distribution across operators.
Rewards Sustainability
Examines the economic incentives provided to operators, including APR/APY rates, reward token characteristics, distribution mechanisms, and sustainability models. The framework favors fee-based reward models over inflationary token emissions, as sustainable rewards generated from user-paid fees provide greater long-term stability and adoption in general.
Project Valuation
Provides insight into market confidence through funding rounds, valuation metrics, and capital raise history. High-valuation projects typically offer more substantial rewards to operators, while investor quality indicates professional validation of the project's prospects.
User Engagement
Measures community interest through social media following, active community participation, content engagement rates, and developer activity. This criterion recognizes that businesses must leverage social media presence to maintain market share and indicates public interest in the product.
Market Fit
Evaluates the project's positioning within its target market, including competitive landscape analysis, timing assessment, and unique value proposition clarity. Congested markets can hinder project traction regardless of technical merit, while clear competitive advantages enhance valuation.
Revenue Streams
Assesses the project's ability to generate sustainable revenue beyond token emissions. Projects with diversified revenue sources and established income streams are more likely to provide stable, long-term rewards to operators.
Strategic Criteria
This category allows for the incorporation of contextual factors not explicitly accounted for in the framework, serving as a "wildcard" for evaluators to assign scores based on unique strategic objectives such as established relationships with specific AVSs.
Relationship
Evaluates the quality of engagement between the AVS project and node operators, including support channel availability, response times, documentation quality, and operator inclusion in governance decisions. Strong operator relationships indicate commitment to long-term collaboration.
Team
Assesses the experience, credibility, and track record of the development team, including prior blockchain experience, successful project history, academic credentials, and industry recognition. Team quality significantly influences project success probability and long-term viability.
This category also accommodates unique strategic considerations such as existing business relationships, financial investments, or specialized expertise that may provide operators with advantages in specific AVS environments.
Practical Application Example
To illustrate the framework's application, consider the following weight distribution for a moderately risk-averse operator:
Category Weights
- Technical: 50% (emphasizing risk mitigation)
- Commitment 25%
- Slashing 25%
- Audits 25%
- Infrastructure 15%
- Software 10%
- Economic: 35% (balanced reward focus)
- TVL 30%
- Rewards 30%
- Valuation 10%
- Engagement 10%
- Market Fit 10%
- Revenue 10%
- Strategic: 15% (standard allocation)
- Relationship 50%
- Team 50%
Sample AVS Evaluation
Now let’s imagine an AVS with the following evaluation yields:
Technical Scores
- Commitment: 95/100 (no additional lockups beyond standard delay)
- Slashing: 75/100 (clear conditions, proportional to Ethereum standards)
- Audits: 85/100 (multiple audits from reputable firms, open-source codebase)
- Infrastructure: 90/100 (standard server requirements, good documentation)
- Software Requirements: 80/100 (containerized deployment, comprehensive guides)
Economic Scores
- TVL: 70/100 (above median for ecosystem, steady growth)
- Rewards: 65/100 (competitive APR, sustainable fee model emerging)
- Valuation: 80/100 (well-funded by reputable investors)
- Engagement: 75/100 (active community, growing developer participation)
- Market Fit: 85/100 (clear value proposition, limited direct competition)
- Revenue: 60/100 (early revenue traction, transitioning from emissions)
Strategic Scores
- Relationship: 80/100 (responsive support, operator-focused governance)
- Team: 90/100 (experienced team with a successful track record)
Final Calculation
Technical Category: (95×25% + 75×25% + 85×25% + 90×15% + 80×10%) * 50% = 42.62
Economic Category: (70×30% + 65×30% + 80×10% + 75×10% + 85×10% + 60×10%) * 35% = 24.67
Strategic Category: (80×50% + 90×50%) * 15% = 12.75
Aggregate Score: 80.04
This score suggests an AVS candidate with strong technical foundations and team credentials, although with moderate economic performance reflecting its early-stage revenue development. The operator would likely include this AVS in their portfolio, potentially allocating a moderate portion of their restaking resources based on their risk tolerance and diversification strategy.
Conclusion
This AVS Evaluation Framework offers a quantitative, structured methodology for assessing AVSs within the EigenLayer ecosystem. By integrating technical, economic, and strategic criteria with adjustable weighting factors, it enables objective comparisons and adaptive decision-making.
As AVSs expand on Ethereum Mainnet, this framework provides a critical tool for evaluating their long-term viability, security, and financial sustainability. Its adoption will promote transparency, informed participation, and enhanced economic security within the EigenLayer ecosystem.
Acknowledgments
This research and framework development was done in collaboration with Bitwise. In particular, we would like to thank Christy, Tim and Chris for their many contributions to this work.
