二月区块链技术更新：Gloria Zhu 辞去比特币核心维护者职务、Solana 路线图、以太坊协议优先级更新GaryMa，Wu BlockchainWuBlockchain 总结了 2 月份区块链技术领域的主要发展：比特币比特币核心开发人员 Gloria Zhao 已辞去比特币核心维护者的职务，并撤销了她的签名密钥。此举不会影响比特币的共识规则、网络安全或交易处理。赵女士主要负责mempool、交易中继等领域，她的离职是开源社区的普遍调整；多重验证显示，她没有参与 2015 年 MIT 相关融资阶段，与 Epstein 相关基金没有直接联系。 比特币社区最近提出了一项新的改进提案 BIP-110，计划通过软分叉限制交易内非货币数据（如 Ordinals 铭文）的大小，以缓解“垃圾数据”占用区块空间的问题。然而，该提议已经引发争议。 Adam Back 表示 BIP-110 可能会损害比特币作为可靠价值存储和安全资产的声誉
，认为其负面影响可能超过所谓的“垃圾交易”本身。 以太坊Glamsterdam升级准备：预计在6月-8月进行升级，保持每年两次升级的节奏； BAL DevNet3 将于 3 月 5 日启动。Hegota 提案提交的截止日期建议延长两周，至 3 月 13 日左右。Vitalik Buterin 概述了以太坊的扩容路线图，强调了 Glamsterdam 分叉中的短期升级，例如块级访问列表、ePBS 和多维 Gas，以更好地管理执行和状态增长，以及以通过 PeerDAS 提高 blob 吞吐量和逐步采用 ZK-EVM 为中心的长期计划，最终移动一旦安全性和形式验证足够，就可以进行多重证明验证
February Blockchain Technology Update: Gloria Zhao Resigns as a Bitcoin Core Maintainer, Solana Roadmap, Ethereum Protocol Priorities Update
Written by | GaryMa, Wu Blockchain
The WuBlockchain summarizes key developments in the blockchain technology space for February:
Bitcoin
Bitcoin Core developer Gloria Zhao has resigned as a Bitcoin Core maintainer and revoked her signing key. This move does not affect Bitcoin’s consensus rules, network security, or transaction processing. Zhao was primarily responsible for areas such as the mempool and transaction relay, and her departure is a common adjustment in open-source communities; multiple verifications show she did not participate in the 2015 MIT-related funding phase and has no direct connection to Epstein-linked funds.
The Bitcoin community recently proposed a new improvement proposal, BIP-110, which plans to limit the size of non-monetary data (such as Ordinals inscriptions) within transactions via a soft fork, to alleviate the problem of “spam data” occupying block space. However, the proposal has already sparked controversy. Adam Back stated that BIP-110 could damage Bitcoin’s reputation as a reliable store of value and a secure asset, arguing that its negative impact may outweigh the so-called “spam transactions” themselves.
Ethereum
Glamsterdam upgrade preparation: the upgrade is expected in June–August, maintaining the cadence of two upgrades per year; BAL DevNet3 will launch on March 5.
The deadline for the Hegota headline proposal submission is proposed to be extended by two weeks to around March 13.
Vitalik Buterin outlined Ethereum’s scaling roadmap, highlighting short-term upgrades in the Glamsterdam fork such as block-level access lists, ePBS, and multidimensional gas to better manage execution and state growth, alongside long-term plans centered on higher blob throughput via PeerDAS and gradual ZK-EVM adoption, eventually moving toward multi-proof validation once security and formal verification are sufficiently mature. — link
Ethereum Foundation researcher Justin Drake released a “strawmap” proposed by the Ethereum Foundation protocol team — an overall technical roadmap draft for researchers and developers — aimed at organizing the upgrade directions for the Ethereum L1 protocol over the coming years from a holistic perspective. The document plans for about seven forks before 2029, envisaging an upgrade cadence of once every six months, and proposes five long-term goals: a faster L1 (second-level finality), 1 gigagas/sec throughput capacity, high-throughput L2 based on data availability sampling, a quantum-resistant cryptographic system, and native private transfer functionality.
The Ethereum Foundation released an update on protocol priorities for 2026, stating that 2026 will advance along three tracks: Scale (integrating L1 execution and Blob scaling), Improve UX (focusing on native account abstraction and cross-chain interoperability), and Harden the L1 (strengthening security, censorship resistance, and network resilience). The update mentions continuing to push the Gas Limit toward 100M and above, advancing ePBS and further Blob parameter increases, promoting the zkEVM attester client, and advancing censorship resistance and post-quantum-related security work; the next major upgrade, Glamsterdam, targets the first half of 2026, with Hegotá planned to follow later within the year.
The Ethereum Foundation (EF) announced the establishment of a brand-new “Platform” team, aiming to optimize the relationship between L1 and L2 so they become a mutually reinforcing system. The team’s work covers protocol development (product-led R&D, incorporating ecosystem needs into the roadmap), protocol integration (providing technical integration guidance for builders and institutions), and strategy and tracking (tracking network health metrics and improving the data presentation of public dashboards). The ultimate goal is to clarify the respective roles of L1 and L2, maximize the advantages of each layer, strengthen Ethereum’s core properties and ETH value through L2 adoption, and provide a clear path for organizational and institutional adoption of Ethereum.
Vitalik Buterin tweeted that combining FOCIL with account abstraction EIP-8141 (based on 7701) creates a synergistic effect: EIP-8141 makes smart accounts (multisig, quantum-resistant signatures, key changes, gas sponsorship) and privacy protocols first-class citizens, allowing operations to be posted directly on-chain without wrappers; FOCIL provides censorship-resistant fast inclusion by randomly selecting 17 executors per slot (proposer + includers); together, smart wallets, gas sponsorship, and private transactions can be included on-chain within almost 1–2 slots even in an adversarial environment; each FOCIL bundle is 8 kB and can be expanded in the future, and most transactions can be included via FOCIL; unlike multi-concurrent proposers (MCP), FOCIL does not control the MEV “last look” role, which is still auctioned by ePBS; even if 100% of slots are bought by adversarial proposers who refuse the public mempool, transactions can still be included quickly, greatly weakening the centralized power of proposers.
Ethereum announced that the ERC-8004 standard has gone live on mainnet. The standard aims to establish a universal identity, reputation, and verification mechanism for AI agents, making AI agents responsible economic participants. The main features of ERC-8004 include: integration with the x402 payment standard for same-layer handling of task evaluation and payments; support for cross-platform agent discovery; portability of reputation to break platform monopolies; on-chain history records to distinguish high-quality agents from scammers; and support for automated reputation checks.
Ethereum co-founder Vitalik Buterin published an article discussing how to achieve 1,000× scaling of L1 by creating new forms of “state.” He pointed out that scaling state is harder than scaling execution and data, and the most practical path is to expand existing state only moderately while introducing an extremely cheap but usage-restricted new type of state. In the future, the existing state tree will mainly be used to store high-value objects (such as core DeFi contracts), while large amounts of user-specific state (such as ERC20 balances and NFTs) will be handled through cheaper but constrained tools. He believes it is entirely feasible to build developer abstractions that are easy to implement for more than 90% of use cases.
Ethereum co-founder Vitalik Buterin tweeted that the original vision of the rollup-centric roadmap is no longer applicable and a new path is needed. Reasons: L2’s progress toward Stage 2 is slow; L1 itself is scaling (low fees, and the gas limit is expected to increase significantly in 2026). The original vision treated L2 as “branded shards” providing fully trusted scaling, but now L2s are unwilling or unable to meet that (partly due to regulatory needs), and L1 has expanded directly. He suggested viewing L2s as a spectrum, with users choosing different degrees of connection based on their needs; L2s should seek unique value beyond scaling (such as privacy, non-EVM VMs, extreme scaling, ultra-low latency, built-in oracles, etc.), handle ETH assets at least to Stage 1, and maximize interoperability. He supports native rollup precompile contracts, leveraging built-in ZK-EVM to achieve EVM verification and synchronous composability without a security council; allowing L2s to self-attest only additional features and clearly inform users of their security guarantees.
Ethereum Foundation member ladislaus.eth wrote that Ethereum is advancing a critical but relatively low-profile architectural shift: moving from nodes repeatedly executing all transactions in a block to confirming execution correctness by verifying zkEVM proofs. The related approach is supported by EIP-8025 (optional execution proofs), allowing validators at the consensus layer to choose to verify multi-client zk proofs instead of running a full execution layer, without requiring mandatory upgrades or forks. This mechanism is expected to significantly reduce node hardware and synchronization costs, benefiting independent validators and home nodes, and laying the groundwork for future L1 execution scaling, ePBS (planned for the Glamsterdam upgrade), and native rollups. The Ethereum L1-zkEVM roadmap targets 2026, and the first dedicated workshop was held on February 11.
Ethereum L2s
Polygon Foundation has announced that the Lisovo hardfork will go live before mainnet block height 83,756,500 (scheduled for March 4 at 14:00 UTC). The upgrade includes subsidizing gas costs for payments between delegation agents (PIP-82), optimizing smart contract compatibility for Count Leading Zeros (CLZ), enhanced support for Passkey wallets, a more flexible fee adjustment mechanism, as well as improved transaction transmission reliability and verification mechanisms. The official recommends that all node operators upgrade Bor to v2.6.0 or Erigon to v3.4.0.
zkSync announced that it will officially deprecate its Lite node and infrastructure support on May 4, 2026, and the ecosystem will continue to integrate around the Era product line. The project stated that this adjustment is intended to concentrate resources and development efforts on the Era mainnet and the expanded ecosystem, make the protocol architecture clearer, and reduce overlapping maintenance and community fragmentation. Lite users and developers will receive migration guidance to ensure a smooth transition.
According to a blog post by the Base engineering team, Base will consolidate the key components that run the network into a self-managed unified codebase, and node operators will need to follow Base release versions rather than directly following Optimism’s versions. In the coming months, Base will gradually detach from OP Stack, but will still cooperate with Optimism as Optimism’s OP Enterprise customer. The project said this move will increase the upgrade cadence from three times per year to six times per year, and in subsequent hard forks will advance an upgrade from optimistic proofs to TEE/ZK proofs, while maintaining Stage 1 rollup status and adding independent signers to the Base security council. Base is the highest-revenue L2 in the OP Stack, which also means that as Base becomes more independent, it may no longer share sequencer revenue with Optimism.
Base issued an explanation regarding the network’s transaction loss and packaging delay issues that occurred on January 31. The project stated that the incident stemmed from a change in transaction propagation configuration, which caused block builders to repeatedly fetch transactions that could not be executed due to rapidly rising base fees. Network stability has now been restored by rolling back the change. Base plans to optimize the transaction pipeline over the next month by eliminating unnecessary P2P overhead, adjusting mempool queues, and improving monitoring and alerting mechanisms during infrastructure changes to prevent similar issues from happening again.
Solana
In an updated technical roadmap on February 3, 2026, the Solana Foundation marked Solana’s formal shift from “pursuing high throughput” to a Solana 2.0 era of “extreme low latency” and “full-stack composability.” Features under development include: Vote Account V4, which aims to optimize the account structure for validator participation in the consensus mechanism by removing unnecessary historical tracking and setting commissions for different types of rewards; Rent Reduction, which will gradually reduce rent deposits required for account creation to lower costs for developers and users; Alpenglow, which introduces a new consensus protocol to shorten confirmation times and simplify how the network reaches consensus on blocks, also introducing the concept of VAT; SIMD-123 enables validators and delegators to share block rewards; 100M CUs will increase block compute unit capacity to boost throughput; SIMD-296 expands transaction size so a single transaction can perform more complex operations; SIMD-268 raises CPI nesting limits, allowing developers to build more complex applications. XDP, as a high-performance networking technology, is available in Agave 3.0+ versions and can accelerate the speed at which validators process network packets.
Hyperliquid
HyperCore will support Outcome contract trading (HIP-4). Outcome is a fully collateralized contract primitive that settles within a fixed range, suitable for prediction markets and bounded derivatives such as option-like products. It involves no leverage or liquidations and features nonlinearity and expiring contracts. This primitive can be combined with portfolio margin and HyperEVM to enhance HyperCore’s expressiveness. Outcome is currently still in the testnet stage; after development is completed, standardized markets based on objective settlement sources will be launched and denominated in USDH, with permissionless deployment planned later based on user feedback.
BNB Chain
BNB Chain announced the launch of its first official BNB Application Proposal, BAP-578. This proposal introduces a new token standard for Non-fungible Agents (NFA), a class of AI-driven assets that can act autonomously on-chain. NFAs can own wallets, execute transactions, maintain an operation history, and operate across different dApps. BAP-578 aims to help developers reach consensus on application-layer standards (such as NFT utility and AI behavior), while laying the foundation for an “agent economy,” transforming AI from a passive tool into an active on-chain entity that can directly hold assets and interact with protocols.
zERC20 announced the official launch of zBNB, enabling private transfers on BNB Chain. zBNB preserves the familiar BNB user experience while using ZK Proof-of-Burn technology to generate cryptographic proofs, fully separating zBNB’s on-chain history from the original BNB transaction records. Users can complete private transfers without additional software: first wrap BNB into zBNB, then send it to the recipient’s burn address; sending BNB directly will result in permanent loss. Binance founder CZ reposted and commented that this is a BNB privacy project developed by a Japanese friend, and stated there is no investment or business relationship.
Security-Related
SlowMist disclosed a supply-chain poisoning incident involving the OpenClaw plugin hub ClawHub. Due to insufficient review mechanisms, a large number of malicious skills were mixed in to spread malicious code or harmful content. Koi Security scanned 2,857 skills and identified 341 malicious skills. SlowMist analyzed 400+ IOC samples pointing to a small number of fixed domains/IPs, and found the attacks exhibited organized and mass-production characteristics; a common two-stage loader was used, with the first stage obfuscated and the second stage dynamically fetching the payload. A typical sample, the “X (Twitter) Trends” skill, concealed a Base64 backdoor that downloaded and executed programs to phish passwords and collected files for upload to a C2 address.
Security Alliance issued a security advisory stating that North Korean (DPRK) IT workers are impersonating real individuals’ LinkedIn accounts to apply for remote positions. These accounts often use the impersonated person’s real information and include verified work emails and identity badges to enhance credibility. Security Alliance recommends that recruiters must verify whether candidates actually control the listed accounts, for example by requiring them to proactively establish contact via LinkedIn. If impersonation is discovered, victims can post warnings on platforms such as LinkedIn, X, or Telegram to prevent more people from being deceived.
Follow us
Twitter: https://twitter.com/WuBlockchain
Telegram: https://t.me/wublockchainenglish