macro-inc/macro
Macro is a unified workspace for teams: email, chat, docs, tasks, agents, calls, and CRM — @-linked together with shared AI memory.
Rust · ★ 5,993 · Fork 546 · 在 GitHub 查看
在手机上浏览源码、订阅每日 AI 趋势日报与项目架构深度解读:获取 CopoHub 客户端(iOS & HarmonyOS)→
GitHub Trending 上榜记录
macro-inc/macro 已在 GitHub Trending 每日榜单上榜 8 天,最高排名第 2,首次上榜 2026-08-12,最近一次上榜 2026-10-31。
| 日期 | 排名 | Star | 当日增长 |
|---|---|---|---|
| 2026-10-31 | #7 | 5,993 | +57 |
| 2026-10-30 | #7 | 5,845 | +379 |
| 2026-10-29 | #5 | 5,614 | +117 |
| 2026-10-28 | #5 | 5,392 | +106 |
| 2026-08-15 | #4 | 3,156 | +436 |
| 2026-08-14 | #4 | 2,995 | +1,239 |
| 2026-08-13 | #2 | 2,561 | +1,180 |
| 2026-08-12 | #2 | 1,677 | +325 |
同期上榜的项目
- thedotmack/claude-mem – Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
- mattpocock/skills – Skills for Real Engineers. Straight from my .agents directory.
- Donchitos/Claude-Code-Game-Studios – Turn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.
- pingdotgg/t3code – The best way to control Claude Code, Codex and more from any computer or phone
- Renhuai123/ziwei-doushu – 紫微斗数开源排盘引擎 — 基于倪海厦《天纪》体系,含完整排盘算法、四化系统、格局知识库、古籍原文数据
- milind-soni/OpenMausBot – Open-source Grok Bot alternative with a virtual machine that bots can use
Macro 项目深度剖析与商业评估报告
Macro 项目深度剖析与商业评估报告
1. 核心定位与商业落地价值
1.1 根本痛点解构 Macro 解决的是现代企业软件“碎片化”的系统性难题。其核心洞察在于:团队协作的真相是信息在工具间流动的断裂——Slack 中的决策、Linear 中的任务、Notion 中的文档、Gmail 中的客户沟通,彼此孤立,导致企业状态(Company State)无法被计算。传统解决方案(Zapier/MCP 拼接)是补丁而非架构,其本质缺陷在于跨系统数据同步的最终一致性无法保证,且缺乏统一的权限模型和数据血缘。
1.2 商业落地场景与价值主张 中小团队 All-in-One 操作系统:面向 5-50 人规模、追求工具极简化的初创公司,替代 4-6 个 SaaS 订阅(Slack+Linear+Notion+HubSpot+Superhuman),TCO 显著降低。 AI Agent 基础设施层:其“团队级记忆”+ 近 100% 工具覆盖率的 MCP 接口,是构建企业级 AI 副驾驶(Copilot)的稀缺底座。相比 LangChain 等框架,Macro 提供的是有状态、有权限、可追溯的 Agent 执行环境。 垂直行业定制:基于 AGPLv3 源码,咨询公司或行业 ISV 可为其构建定制化 CRM/工单系统,利用其双向图模型快速搭建行业解决方案。
1.3 竞品差异化与护城河 | 维度 | Macro | Notion+Slack+Linear 组合 | Attio/HubSpot | |---|---|---|---| | 数据模型 | 统一双向图 | 多系统孤立 | 单一 CRM 域 | | AI 记忆 | 团队级、跨模态、夜间合成 | 无/单聊天记录 | 仅 CRM 数据 | | 技术栈 | Rust+CRDT 高性能 | 各系统异构 | 闭源 |
其真正的护城河并非单一功能,而是“双向链接 + 统一权限 + 共享记忆”三者叠加产生的网络效应:数据越丰富,AI 记忆越准确,Agent 越有用,用户粘性越强。这是典型的 Data Network Effect。
2. 核心架构设计与技术深度
2.1 模块化单体(Modular Monolith)与微服务混合架构 仓库包含 42 个 deployable services 和 167 个 Rust crates,表面是微服务,实则是以领域为核心的高内聚模块化单体。其服务边界遵循 Hexagonal Architecture(六边形架构),每个服务内部严格区分: Inbound Adapters:HTTP/gRPC/MCP 接口层 Domain Core:纯业务逻辑与领域事件 Outbound Adapters:Postgres/Redis/对象存储等基础设施
在 README 中展示上榜徽章
[](https://github.fq6825.top/repo/macro-inc/macro)