https://chainte.github.io/ckb-talk-radar/ Daily Nervos Talk community briefing and latest active topics. zh-cn Wed, 29 Apr 2026 02:01:11 +0000 https://chainte.github.io/ckb-talk-radar/ tag:ckb-talk-radar,2026-04-29:daily-brief Wed, 29 Apr 2026 03:22:32 +0000 ## 今日发生了什么 今天论坛的核心动态集中在技术基建与星火计划项目推进上：CKB-VM 获得了新的差分测试框架，一个去中心化众筹项目上线测试网自动结算机器人，而多个星火计划资助项目进入了评审、周报更新或收尾阶段。[S01, S02, S04, S05, S08] ## 重点话题 **CKB-VM 安全基建双线并进**：CKBdev 团队实现了 CKB-VM 的差分测试框架，用于验证优化和未来架构变更；与此同时，社区成员提交的 CKB-VM Sail 形式化验证提案因超出星火计划"低门槛、快节奏"的资助范围，被建议调整后再正式评审。[S01, S02] **CKB Kickstarter 测试网实现完全无信任化**：Ayoub_Lesfer 更新了自动结算机器人已在测试网部署并完成端到端验证，众筹流程（创建→认捐→截止→结算）现在无需人工介入即可自动完成。[S04] **星火计划项目状态分化**：Dular 项目因需补充 Daraja 生产环境凭据和修正 CKB 发放机制表述，被暂定为 Pending；Nervos Brain 发布第七周周报，重点推进工具闭环、评测基线和 Telegram/Discord 稳定性，并计划本周拉试用群邀请委员会体验。[S03, S05, S06] **DAO v1.1 项目正式收尾**：_magicsheep 确认项目终止，已交付的 Milestone 1 对应款项保留，代码继续开源。[S08] **Cellora 索引服务获社区反馈**：ArthurZhang 参与讨论，建议交易包含证明优先利用 CKB 现有 RPC 接口而非直接采用... https://talk.nervos.org/t/teamckb-dev-log-updated-apr-29-2026/8572 https://talk.nervos.org/t/teamckb-dev-log-updated-apr-29-2026/8572#recent-1 Wed, 29 Apr 2026 02:01:11 +0000 Updates Features Differential Testing Framework for ckb-vm Implemented a differential test framework for ckb-vm. This provides a stronger foundation for validating optimizations and future architecture changes: GitHub - yuqiliu617/ckb-vm-contrib at differential-test · GitHub CKB DAO Treasury Design & Voting Research research on the CKB DAO Treasury design and voting settlement, with multiple directions under active discussion: Option 1: DAO-Bound Voting with Off-Chain State Explores a governance model where voting power is bound to Nervos DAO deposits, while proposal state, voting records, and tally data are maintained off-chain and committed on-chain through verifiable state roots. This... https://talk.nervos.org/t/spark-program-ckb-vm-sail-formal-verification-proving-ckb-vm-risc-v-instruction-equivalence-via-sail-specification-and-coq-theorem-prover-ckb-vm-sail-sail-coq-ckb-vm-risc-v/10214 https://talk.nervos.org/t/spark-program-ckb-vm-sail-formal-verification-proving-ckb-vm-risc-v-instruction-equivalence-via-sail-specification-and-coq-theorem-prover-ckb-vm-sail-sail-coq-ckb-vm-risc-v/10214#recent-1 Wed, 29 Apr 2026 02:00:41 +0000 @TinyuengKwan 你好，感谢提交 ckb-vm-sail-verify 的提案。这个方向非常关键：CKB-VM 是 CKB 的执行层安全基石，而你提出的 “Sail（官方规范）+ Coq（形式化证明）+ 差分测试（工程验证）” 的双轨方案，也能看出你在 Sail/ACT 生态中有相对稀缺的一手经验与工程积累。 需要说明的是，该项目已明显超出星火计划的支持范围（以 低门槛、快节奏 的方式帮助社区开发者启动小型原型项目）。如果你仍希望向委员会正式递交该项目，那么在提交委员会正式评审前，我个人建议你按 Spark 的预审要求补充/澄清以下几点信息，让评审能更快收敛到“本期资助是什么、怎么验收、验收失败影响什么”。 1) 验证方案（How to Verify）——写成“可复现步骤 + 通过/失败标准 + 影响面（blast radius）” 你在主贴里讲清楚了技术架构与实现设想，但仍缺少一段面向评审者/社区的 “低成本、可重复” 的验收章节。建议新增一个独立章节 How to Verify（或在 repo 内提供 VERIFICATION.md），至少包含： 1.1 可复现步骤（评审者照着做即可） 建议最少做到以下级别： 环境一键化：提供 Docker/Nix/脚本，确保在干净 Linux 环境下可以复现（并 pin 关键依赖版本：Sail/Coq/OPAM/Rust 等）。 生成与构建：从 sail-riscv 生成 Coq 产物的命令（以及预期输出在哪里）。 证明检查：如何运行 Coq proof check（命令 + 预期输出）。 差分测试：如何运行 diff-... https://talk.nervos.org/t/spark-program-dular/10212 https://talk.nervos.org/t/spark-program-dular/10212#recent-1 Wed, 29 Apr 2026 01:49:03 +0000 @duongja 你好，感谢你提交 Dular 项目提案，并根据预审建议补充了 How to Verify、预算拆分与风险说明等内容。整体方向（Fiber + RUSD/UDT + 真实在地试点）很契合 Spark 对“可落地、可验证”的资助导向。 本项目当前状态暂定为 Pending，并非否定项目价值，而是表示：在进入下次正式评审/投票前，我们还需要你补齐两类“可验证凭据”，并纠正提案中关于 CKB 发放与汇率折算机制的表述，以避免后续沟通成本与验收争议。 1) 请补充：Daraja 生产环境凭据的可视化证据。 你已声明持有 STK Push + B2C 生产凭据，请通过私信发我一张 Daraja Developer Portal 的后台截图（API Key 字段可全部打码，但需保留 Production 环境标识、App 名称、创建时间）。这一步用于委员会内部尽调，不会公开。委员会在验证后，会在本帖中发布一条信息，说明已经核验。 2) 请补充：现有 multi-hop RUSD payment 的可核验证据。 为了让委员会/社区能低成本持续复核你的技术交付（而不是只在结项时“补材料”），请在主楼直接给出对应的 Fiber payment hash 或相关 CKB testnet tx hash，并简要说明 Fiber payment hash 与 CKB L1 explorer 的对应关系（避免读者误以为每笔 Fiber 支付都能在 CKB explorer 直查）。 3) 关于 CKB 发放与汇率风险：请修正提案中的流程假设，并按 Spark 口径重写 你在提案的... https://talk.nervos.org/t/introducing-ckb-kickstarter-decentralized-all-or-nothing-crowdfunding-on-nervos-ckb-testnet-mvp-live/10130 https://talk.nervos.org/t/introducing-ckb-kickstarter-decentralized-all-or-nothing-crowdfunding-on-nervos-ckb-testnet-mvp-live/10130#recent-1 Tue, 28 Apr 2026 17:10:24 +0000 Update: Automatic Finalization Bot live on testnet Following up on the v1.1 update above: the bot is deployed and end-to-end verified on testnet as of yesterday (2026-04-27). The platform is now fully trustless on testnet, campaigns flow create → pledge → deadline → distribution with zero manual intervention from anyone (creator, backer, or platform operator). What the bot does (each polling cycle, every 10s): Detects expired campaigns still in Active status → submits permissionless finalizeCampaign tx (Success if total pledged ≥ goal, Failed otherwise) For finalized Success campaigns with remaining live pledge cells → submits permissionlessRelease tx (funds → creator) For finalized... https://talk.nervos.org/t/spark-program-nervos-brain-a-global-developer-onboarding-engine-and-cross-language-hub-powered-by-agentic-rag/9995 https://talk.nervos.org/t/spark-program-nervos-brain-a-global-developer-onboarding-engine-and-cross-language-hub-powered-by-agentic-rag/9995#recent-3 Tue, 28 Apr 2026 14:03:27 +0000 第七周周报 一、本周目标（工具闭环与评测基线周） 本周承接第六周“多轮可持续交互”阶段的工作，重点从“机制已经具备”推进到“关键路径真正闭环、且后续可以被稳定评测”。核心目标有四个： 继续治理运行时日志噪音，补齐最小可观测性闭环。 让 discourse_query / github_search 从协议层定义走到图执行主路径可调用。 建立第一版多轮评测集，为后续 benchmark 和量化回归提供统一输入。 补齐 Telegram / Discord 两端在长消息与异常路径下的稳定性回归。 二、本周完成 日志治理与诊断视图补齐 已对常见第三方库日志进行了统一降噪处理，补充了 quiet_loggers 与 third_party_level 控制项，降低了测试和运行期的无关日志干扰。同时把工具执行过程的摘要信息接入到图状态与 trace_summary 中，使得一次回答至少可以追踪到“执行了哪些 tool、各自成功/为空/失败”的最小诊断视图，而不再只能看到最终回答文本。 工具执行闭环推进到运行时主路径 本周把 discourse_query 与 github_search 从“schema 已存在但主路径未接通”的状态推进到了可被 RetrieverPlanner -> RetrievalExecutor -> ToolRuntime 实际调用的状态。具体包括： RetrieverPlanner 归一化阶段已允许这两个 tool 保留，不再强制回退到 qdrant_search。 ToolRuntime 为二者补齐了 handler。 handler... https://talk.nervos.org/t/discontinuation-of-the-dao-v1-1-project/10204 https://talk.nervos.org/t/discontinuation-of-the-dao-v1-1-project/10204#recent-1 Tue, 28 Apr 2026 11:56:27 +0000 In consideration of Terry’s advice, the following updates are provided regarding the closure of DAO v1.1: Payment: The proposal team will retain the payment corresponding to the already‑delivered Milestone 1. Code: The code will remain open source and is accessible at this repository. A total of eight repositories encompass all code for the DAO v1.1 platform (excluding Web5 services, which are available here). Additionally, the community may access the open‑source vote auditor tool here. Contract: The voting contract deployed on the mainnet has been terminated, whereas the testnet voting contract continues to operate. The did:ckb contracts remain active on both the mainnet and the... https://talk.nervos.org/t/cellora-designing-a-production-indexing-and-query-service-for-ckb-feedback-welcome/10199 https://talk.nervos.org/t/cellora-designing-a-production-indexing-and-query-service-for-ckb-feedback-welcome/10199#recent-1 Tue, 28 Apr 2026 08:40:25 +0000 Just came across this thread and found it interesting, so I’ll try to offer a few suggestions. I think the honest answer is: For tx inclusion proofs, the practical first step is likely not Flyclient, but exposing CKB’s existing get_transaction_proof / verify_transaction_proof path through Cellora. That lets clients verify that a transaction is committed under a particular block header, rather than merely trusting Cellora’s indexed result. This easily moves Cellora from a purely trusted indexer toward an inclusion-verifiable indexer. For full historical / chain-tip trust minimisation, my narrower point is that I’m not sure there is a canonical Rust/TS wallet-side verifier package that app...