Ethereum Foundation Protocol Cluster Releases Unified Tier List for Hegoté Upgrade

The Ethereum Foundation (EF) Protocol cluster has officially published its first-ever unified tier list for the upcoming Hegoté network upgrade. Evaluating a total of 62 Ethereum Improvement Proposals (EIPs) proposed for inclusion, the comprehensive assessment provides clear insight into which upgrades are locked in, which remain under consideration, and which have been definitively declined.

For the first time in the history of these evaluations, the cluster has abandoned individual per-team opinions in favor of a single, unified view. Approximately 60 people spanning research and engineering across nine core teams, along with several individual domain experts, collaborated to produce the definitive rankings. While Execution Layer client Geth will still release its own standalone tier list, the broader Protocol cluster integrated feedback from every participating team and individual contributor to establish a single source of truth.

The grades and priorities guiding the Hegoté upgrade draw directly from a companion publication titled EF Protocol: Current and Emerging Priorities, which outlines the cluster’s overarching north star, strategic roadmap, and core commitments.

How the Tiering Process Worked

The evaluation process relied on extensive collaboration, independent scoring, and rigorous debate. Sixteen contribution templates were completed across the nine participating teams and external domain experts. Twelve of those contributors provided tier grades, with several focusing exclusively on EIPs within their areas of deep technical expertise. In total, the community generated 397 individual tier grades across the 62 proposals, averaging 6.4 grades per EIP, with the most heavily debated items drawing up to nine independent assessments.

The scoring methodology treated each tier as a distinct commitment level. Independent contributors graded proposals on a numeric scale where S represented 4 points, A equaled 3, B was 2, C scored 1, and D or DFI (Deferred for Inclusion) registered as 0. Final averages factored in only cast grades, ensuring that abstentions never penalized a proposal.

Lessons learned from previous retrospectives on the Glamsterdam upgrade heavily influenced the methodology. Reviewers noted that an EIP’s true size should be measured by its integration depth, that technical complexity compounds rapidly, that the available testing surface remains a scarce resource, and that proposal champions frequently underestimate the implementation complexity of their ideas. To finalize the list, the cluster held half a dozen group calls and an intensive 90-minute working session to debate contested items, resulting in a finalized steelman view where individual per-team grades were kept private.

Consensus Layer Proposals: Headliners and Deferred Items

Within the Consensus Layer, proposals were sorted into strict performance and readiness tiers. EIP-7805, known as FOCIL, emerged as the undisputed headliner, capturing a unanimous S-tier rating under full participation. FOCIL establishes a crucial mechanism allowing any user to include an eligible transaction without depending on centralized builders. Protocol developers emphasize that FOCIL should ship alongside EIP-8369, which defines transaction eligibility criteria and validator verification rules, effectively extending inclusion guarantees to Frames.

Several other consensus proposals secured strong A-tier positions. EIP-8015 secured a near-unanimous rating for cleanly removing dead deposit and eth1data fields that lack downstream dependencies. EIP-8365 introduces the early retirement of withdrawal credentials tied to vulnerable cryptography, acting as the primary post-quantum readiness item deemed appropriate for Hegoté. EIP-8383 reduces the consensus layer block retention window, while EIP-8334 introduces bundled attestation propagation to cut gossip network load. EIP-8025 also landed in the A-tier, upstreaming stateless-execution changes into canonical execution specifications to unify zero-knowledge virtual machine development.

Other consensus changes faced hurdles. EIP-8198, proposing "Quick Slots," received strong theoretical support from research and development teams but encountered sharp pushback from engineering delivery teams. Citing the complex cascade of adjustments required for any change to slot timings, the cluster established four strict prerequisites before an A-tier evaluation can resume: a complete specification covering all core protocol changes, a working prototype implementing that specification, an in-depth ecosystem assessment of downstream effects, and formal sign-off confirming that the change does not complicate decoupled consensus designs.

Numerous consensus proposals were designated as DFI (Deferred for Inclusion). These included EIP-7716 regarding anti-correlation attestation penalties, EIP-8367 concerning balance sunsets for retired BLS validators, and EIP-8243 for batching attestations at the source. Proposals like EIP-8142, which introduced Block-in-Blobs, and EIP-8148, focused on custom sweep thresholds for validators, were unanimously pushed below the line due to excessive dependencies or a failure to meet the strict necessity bar for the Hegoté fork.

Execution Layer Enhancements and Account Abstraction

On the Execution Layer, EIP-8141 emerged as a dominant S-tier headliner. Introducing Frame Transactions, the proposal serves as a cornerstone for native account abstraction rooted in security. It establishes a viable path toward post-quantum signature schemes without requiring a separate fork for each new algorithm, implements transaction aggregation to properly price post-quantum verification, and creates a clear route toward retiring legacy keys. EIP-8141 is slated to ship alongside EIP-8250, which introduces keyed nonces for anonymity and shared senders, and EIP-8272, which leverages recent on-chain state roots for private transactions.

A series of robust A-tier proposals round out the Execution Layer scope. EIP-3298 removes legacy refund mechanisms to eliminate metering edge cases that complicated past forks. EIP-5920 introduces the PAY opcode, offering a low-surface, high-leverage method for transferring value without invoking recipient code. EIP-8279 and EIP-831-applied as a bounding pair-target block access list byte floors and unified transaction content floors to bound worst-case block construction and secure network operations against adversarial congestion.

Account abstraction and key management extensions also found strong support in the A-tier. EIP-7906 introduces transaction assertions via a state diff opcode to prevent wallet drains and MEV extraction. EIP-8298 and EIP-8151 form a coordinated package to retire legacy master keys, allowing EIP-7702-delegated accounts to transition into true smart contract accounts while restricting vulnerable authentication checks. Meanwhile, EIP-4758 successfully overcame its lower average score to secure an A-tier override, supported by engineering teams that maintain the EVM and would otherwise bear the ongoing testing and maintenance cost of the legacy SELFDESTRUCT opcode indefinitely.

Conversely, numerous execution proposals were deferred or declined. EIP-8163, which proposed reserving the EXTENSION opcode, was placed in the DFI block because no committed consumer requires it for this upgrade cycle. Similarly, EIP-2488 to deprecate CALLCODE and EIP-7645 to alias ORIGIN to SENDER were unanimously declined due to application-layer breakage risks and a lack of immediate upgrade urgency.

Conclusion and Next Steps

Out of the 62 proposals rigorously evaluated by the EF Protocol cluster, only two have been designated as must-ship headliners, while 15 are expected to make the final cut for Hegoté. A total of 28 proposals have been officially declined with explicit justifications provided. The remaining proposals consist of conditional B-tier items with precise inclusion requirements, C-tier proposals sitting below the line, and a small number of TBD items awaiting concrete mainnet data.

Cluster members stress that the success of the Hegoté upgrade will depend as much on the proposals they choose to exclude as on those they successfully deploy. To facilitate open discussion and address community feedback, the Protocol cluster will host a public Reddit AMA on r/ethereum. Proposal champions, developers, and community members are encouraged to submit questions ahead of time, with a particular welcome extended to advocates of proposals that did not secure an A-tier ranking or were ultimately deferred.