最新资讯
dev.to 在信任 AI 生成的测试用例前,如何逐条复核 unite.ai Franklin Templeton 构建 AI 研究工作流以揭示风险 techxplore.com AI 助力寻找更优的钛合金 3D 打印工艺参数 engadget.com Perplexity 推出“teammates”日常工作任务 AI agent computerworld.com 别提数据中心:OpenAI 改称“compute campuses” theguardian.com 俄美试图阻止就自主武器制定具有约束力的规则 bloomberg.com Nvidia 加强东南亚芯片追踪以打击走私 thestack.technology 模型路由需求升温,Agent Router 推出开源替代方案 www.youtube.com Google DeepMind 的 Philipp Schmid:AI 同事需要具备的七个要素 unite.ai Cashfree 为 AI Agent 的托管结账加入 WebMCP 工具 dev.to 我们在 DolphinBench 上测试记忆检索:纯关键词搜索暴露了自身回归 dev.to 别再把 Agent 写成 while 循环,把它建模为状态机
多元动态
登录 注册
AI编码 news.ycombinator.com ·

Show HN:Foremerge——捕获并行编码 Agent 之间的意图冲突

原文 我的收藏

Foremerge 是一个位于 git 之上的本地协调层,不改变 git 与 worktree 的工作方式。每个 agent 在编辑前发布 intent 及其将变更的 scope 和操作类型,例如一个 agent 声明 replace PaymentService、另一个声明 extend PaymentService 时,第二次发布会在写代码前返回 HIGH 级 destructive_vs_additive 冲突。共享状态是 git common 目录下的一个 SQLite 文件,无需 hooks 或 merge driver。它以单个 Rust 二进制发布,含 CLI 和 18 个工具的 MCP server,可通过 foremerge setup all 接入 Claude Code、Codex 和 Cursor,协议不绑定特定厂商。检测是确定性的,不依赖评审模型;HIGH 冲突仅针对已声明的操作,从自然语言推断的匹配不会达到 HIGH。声明是建议性租约而非锁,两个 agent 可持有同一 scope 而不死锁。作者称已测试最多 98 个并行 agent 在同一仓库工作且无冲突,并回放 76 条 intent 发现 1 个冲突,同时暴露一个盲点:一个 agent 按类名声明 scope、另一个按内部方法声明。开源版为单机匹配,不是分布式共识。

AI推荐理由 Show HN 首发项目,作者给出具体机制(intent publish + scope 声明、SQLite 共享状态、确定性检测、MCP 18 工具、接入 Claude Code/Codex/Cursor),并披露 98 个并行 agent 测试与 76 条 intent 回放中 1 个冲突及类名/方法名作用域盲点,信息密度高且局限明确。但证据均为作者自述,无第三方验证,98 agent 测试与 76 条回放样本量小,且开源版为单机匹配、非分布式共识,故质量分未给高分。

72

正文

https://github.com/naw103/foremerge Show HN: Foremerge – Catch Intent Conflicts Between Parallel Coding Agents (github.com/naw103 ) At, GPTree, we run several coding agents across our team on one repo using parallel worktrees. Apart from wasted time reviewing and fixing conflicts at PR time, the failures that hurt the most are when multiple plans or tickets cause architecture changes that cannot both be true. Ex. one agent replaces a class while another one is in the process of extending it. Git only notices if the resulting patches happen to touch the same lines and the review only catches it if they are familiar with both tickets.Foremerge is a local "git like" coordination layer that sits above git (ie. does not interact with or change the way git and worktrees function), Before editing each agent publishes an intent and the scopes it will change, with the operation it plans to complete on each one. foremerge intent publish --agent "$A" \ --summary "Replace PaymentService with StripePaymentService" \ --scope symbol:PaymentService=replace foremerge intent publish --agent "$B" \ --summary "Add PayPal support to PaymentService" \ --scope symbol:PaymentService=extend The publish by the 2nd agent returns a HIGH destructive_vs_additive finding before writing any code. Agents keep their own worktrees and the shared state is one SQLite file in gits common direectory. No hooks, no merge drivers, nothing rewrites your history.It ships as one Rust binary with a CLI and MCP server with 18 tools and `foremerge setup all` wires it into Claude Code, Codex and Cursor. Because the protocol has nothing provider specific, a Claude agent and a Codex agent coordinate through the same store. Before any work is accepted, Foremerge runs a named check that you configured against the exact git state of the change. An agent that says tests pass is recorded but it dosnt satisfy the acceptance gate without running the check itself.Detection is deterministic, no judge model reading your code. HIGH conflicts are only asserted for declared operations, ie. matches inferred from prose cap out below high. Claims are advisory leases, not locks so two agents can still hold the same scope without deadlock. The open source version is single matching and so not a distributed consensus.We have tested this up to 98 parallel agents all working on the same repo with zero conflicts (was supposed to be 100 but 2 agents failed to run due to resource limitations)I replayed 76 intents on my own agents from a build last week in the order they happened. The sample had exactly 1 conflict (which was flagged) and the review found a blind spot where one agent claimed scope by class name and the other claimed it by an internal method. We are working on fixing that for the next release.Setup is a 30s install by pasting the quickstart instructions from the readme.md into your agent or manually: `curl -fsSL https://foremerge.com/install.sh https://foremerge.com/install.sh | sh` or `cargo install --locked foremerge`, then `foremerge init && foremerge setup all` in a repo. Apache-2.0.The feedback I want most is which conflicts between your agents plans would you actually want flagged and which would you tollerate as noise?Repo here: https://github.com/naw103/foremerge https://github.com/naw103/foremerge Website: https://foremerge.com https://foremerge.com More information on the problems this solves: https://foremerge.com/blog/ https://foremerge.com/blog/ help Guidelines | FAQ | Lists | https://github.com/HackerNews/API API | Security | https://www.ycombinator.com/legal/ Legal | https://www.ycombinator.com/apply/ Apply to YC | Contact Search:

来源报道 · 1 篇 · 以下链接均前往原始媒体

← 返回精选动态 原文

更多内容