AI that thinks, acts, remembers — and keeps going when one model dies.
+Your AI needs a Plan B.
[](https://github.com/matevip/mateclaw) [](https://claw.mate.vip/docs) @@ -28,9 +28,9 @@ --- -Most AI tools forget you the moment the tab closes. Most fall over when their model vendor has a bad day. Most give you a chatbox and call it a product. +Most AI tools die when their vendor has a bad day. Most forget you the moment the tab closes. Most give you a chatbox and call it a product. -**MateClaw is the whole widget.** One deployment. Reasoning, knowledge, memory, tools, and multi-channel presence — built together, not bolted on. +**MateClaw is the whole widget.** One deployment. Reasoning, knowledge, memory, tools, channels — built together, not bolted on. And when your primary model goes down, the next one picks up mid-sentence. --- @@ -42,7 +42,7 @@ Primary key expired. Vendor returns 401. Network blip. Quota drained. Other tools hand you a red error card. MateClaw routes to the next healthy provider — DashScope, OpenAI, Anthropic, Gemini, DeepSeek, Kimi, Ollama, LM Studio, MLX, 14+ in total — and the user sees the reply finish. A provider health tracker parks bad vendors in a cooldown window so they don't waste seconds on every turn. -You don't configure a retry script. You set priorities in the web UI. The runtime does the rest. +You don't write a retry script. You drag providers into priority order in **Settings → Models** and watch the health dashboard fill with green dots as requests route around failures in real time. ### 2 · Knowledge that links itself @@ -64,6 +64,8 @@ This is the difference between a warehouse and a library. Same brain. Same memory. Same tools. Different doors. +$0 · No tokens metered. No seats billed. Your server. Your data. Your keys.
+ --- ## What's in the box @@ -87,17 +89,32 @@ RBAC + JWT. Full audit trail. Flyway-managed schema that auto-heals on upgrade. --- +## AI is becoming infrastructure + +On March 2, 2026, Claude went dark for 4 hours across API, web, and mobile. Three weeks later, another 5 hours. Every company that bet their AI strategy on a single vendor spent those outages staring at red error cards. + +This is the same shift databases went through around 2010 and cloud went through around 2018: the winning layer stops being tied to one supplier. **57% of companies now run AI agents in production.** None of them want one vendor's bad day to become their bad day. + +**MateClaw is that layer — built the Spring Boot way.** + +--- + ## Why MateClaw -| | MateClaw | Claude Code | Cursor | Windsurf | -|:---|:---:|:---:|:---:|:---:| -| **Multi-model failover** | Auto-route across vendors | Anthropic only | One model | One model | -| **Knowledge digestion** | Wiki with citations | CLAUDE.md only | Code index | — | -| **Multi-channel presence** | 7 IM + Web + Desktop + Widget | 3 IM preview | IDE only | IDE only | -| **Admin dashboard** | Full web console | Enterprise tier | — | — | -| **Price** | **Free · Apache 2.0** | $20–200/mo | $0–200/mo | $0–200/mo | +| | MateClaw | [OpenClaw](https://github.com/openclaw/openclaw) | [Hermes Agent](https://github.com/NousResearch/hermes-agent) | [Claude Code](https://github.com/anthropics/claude-code) | [Cursor](https://cursor.com) | +|:---|:---:|:---:|:---:|:---:|:---:| +| **Multi-vendor failover** | **Chain + health tracker + cooldown** | Swap providers via config | Orchestration w/ retry | Anthropic only | One model | +| **Knowledge digestion** | **LLM Wiki + page-level citations** | Canvas + memory | Skills Hub + memory | — | Code index | +| **Multi-user admin** | **RBAC + approval flow + audit** | Config-file first | Single-user CLI | Enterprise tier | Teams plan | +| **Surfaces** | Web admin + Desktop + Widget + SDK + 6 IM | 25+ chat channels | 15+ channels (CLI-led) | 3 IM preview | IDE only | +| **Stack** | **Java (Spring Boot)** | TypeScript | Python | TypeScript | Electron/TS | +| **License / Price** | **Apache 2.0 · Free** | MIT · Free | MIT · Free | Proprietary · $20–200/mo | Proprietary · $0–200/mo | -The comparison table everyone writes is the one that flatters themselves. This is the one that matters: in a category crowded with coding assistants, MateClaw's bet is **generality** — an AI operating system, not an IDE plugin. +**OpenClaw and Hermes Agent are excellent personal AI platforms** — pick either if you're running one user on one laptop, building your own agent from CLI, and treating everything as config files to hand-tune. Both have bigger communities than MateClaw today. + +**MateClaw is the version built for teams.** RBAC per agent, per model, per tool. An approval flow that pauses risky actions for review. Full audit trail. A web admin dashboard where one operator manages 50 agents across 14 vendors. Spring Boot inside — drop-in for any Java shop already running production services. + +Same "whole widget" philosophy. Different center of gravity. --- @@ -149,16 +166,15 @@ Download from [GitHub Releases](https://github.com/matevip/mateclaw/releases). B mateclaw/ ├── mateclaw-server/ Spring Boot 3.5 backend (Spring AI Alibaba, StateGraph runtime) ├── mateclaw-ui/ Vue 3 + TypeScript admin SPA (built into the server JAR) -├── mateclaw-desktop/ Electron app with bundled JRE 21 ├── mateclaw-webchat/ Embeddable chat widget (UMD / ES bundles) ├── mateclaw-plugin-api/ Java SDK for third-party capability plugins ├── mateclaw-plugin-sample/ Reference plugin implementation -├── matevip-sites/ Marketing site + VitePress docs (pnpm workspace) -├── deploy/ Production deployment configs ├── docker-compose.yml └── .env.example ``` +Desktop binaries ship via [GitHub Releases](https://github.com/matevip/mateclaw/releases) with a bundled JRE 21 — no Java install needed. + ## Tech stack | Layer | Technology | diff --git a/README_zh.md b/README_zh.md index a0611a26..5a81e5e5 100644 --- a/README_zh.md +++ b/README_zh.md @@ -6,7 +6,7 @@ # MateClaw -会思考、会行动、会记忆的 AI——主模型挂了，它还能继续干活。
+AI 也该有 Plan B。
[](https://github.com/matevip/mateclaw) [](https://claw.mate.vip/docs) @@ -28,9 +28,9 @@ --- -大多数 AI 工具一关标签页就忘了你是谁。供应商一抽风就两手一摊。给你一个聊天框，就敢叫产品。 +大多数 AI 工具一到厂商抽风那天就两手一摊。关一次标签页就忘了你是谁。给你一个聊天框，就敢叫产品。 -**MateClaw 是完整的一整套。** 一次部署——推理、知识、记忆、工具、多渠道入口，从第一天就是一起设计的，不是事后拼接的。 +**MateClaw 是完整的一整套。** 一次部署——推理、知识、记忆、工具、多渠道入口，从第一天就一起设计，不是事后拼接。主模型挂了，下一家接着把这句话说完。 --- @@ -42,7 +42,7 @@ Key 过期。厂商返回 401。网络抖动。配额耗尽。 别的工具丢你一张红色错误卡。MateClaw 自动切到下一家健康的供应商——DashScope、OpenAI、Anthropic、Gemini、DeepSeek、Kimi、Ollama、LM Studio、MLX，共 14+ 家——用户只会看到回答正常完成。内置的 **Provider Health Tracker** 会把连续失败的供应商放进冷却窗口，避免每一轮对话都白白撞壁。 -你不用写重试脚本。在 Web UI 里设定好失败转移优先级，运行时自己搞定。 +你不用写重试脚本。在 **设置 → 模型** 里把供应商拖成你想要的优先顺序，健康面板实时亮起一排绿点——请求绕着故障流过去。 ### 2 · 知识会自己长出链接 @@ -64,6 +64,8 @@ MateClaw 的 **LLM Wiki** 把它消化成结构化页面，页面之间自己长 同一个大脑。同一份记忆。同一套工具。不同的门。 +$0 · 无 token 计费。无座位收费。你的服务器，你的数据，你的 Key。
+ --- ## 盒子里有什么 @@ -87,17 +89,32 @@ RBAC + JWT。完整审计事件流。Flyway 管理数据库 schema，升级时 --- +## AI 正在变成基础设施 + +2026 年 3 月 2 日，Claude 全球宕机 **4 小时**——API、Web、移动端同时黑屏。三周后又来一次，**5 小时**。每一家把 AI 战略押在单一厂商身上的公司，那几个小时只能盯着红色错误卡。 + +这和 2010 年数据库走过的路、2018 年云走过的路**是同一个转弯**：赢的那一层，不再绑在一家供应商身上。**57% 的公司已经把 AI agent 推进生产**——没有一家希望某个厂商的坏日子变成自己的坏日子。 + +**MateClaw 就是那一层——用 Spring Boot 方式盖的。** + +--- + ## 为什么选 MateClaw -| | MateClaw | Claude Code | Cursor | Windsurf | -|:---|:---:|:---:|:---:|:---:| -| **多模型自动切换** | 跨厂商自动路由 | 仅 Anthropic | 单模型 | 单模型 | -| **知识消化式加工** | Wiki + 引用溯源 | 仅 CLAUDE.md | 代码索引 | — | -| **多渠道入口** | 7 IM + Web + 桌面 + 嵌入 | 预览 3 渠道 | 仅 IDE | 仅 IDE | -| **管理仪表盘** | 完整 Web 控制台 | 企业版 | — | — | -| **价格** | **免费 · Apache 2.0** | $20–200/月 | $0–200/月 | $0–200/月 | +| | MateClaw | [OpenClaw](https://github.com/openclaw/openclaw) | [Hermes Agent](https://github.com/NousResearch/hermes-agent) | [Claude Code](https://github.com/anthropics/claude-code) | [Cursor](https://cursor.com) | +|:---|:---:|:---:|:---:|:---:|:---:| +| **多厂商失败转移** | **Chain + 健康追踪 + 冷却** | 切换供应商（改配置） | 内置编排重试 | 仅 Anthropic | 单模型 | +| **知识消化式加工** | **Wiki + 页面级引用溯源** | Canvas + 记忆 | Skills Hub + 记忆 | — | 代码索引 | +| **多用户管理** | **RBAC + 审批流 + 审计** | 配置文件优先 | 单用户 CLI | 企业版 | 团队版 | +| **用户触点** | Web 管理台 + 桌面 + 嵌入 + SDK + 6 IM | 25+ 聊天渠道 | 15+ 渠道（CLI 为主） | 3 IM（预览） | 仅 IDE | +| **技术栈** | **Java（Spring Boot）** | TypeScript | Python | TypeScript | Electron/TS | +| **许可 / 定价** | **Apache 2.0 · 免费** | MIT · 免费 | MIT · 免费 | 闭源 · $20–200/月 | 闭源 · $0–200/月 | -每个产品写的对比表都是对自己有利的。这张表要说的是：在编码助手扎堆的赛道里，MateClaw 下的注是**通用性**——AI 操作系统，不是 IDE 插件。 +**OpenClaw 和 Hermes Agent 是优秀的个人 AI 平台**——如果你是一个人、一台笔记本、习惯从 CLI 搭自己的 agent、所有东西都靠手工配置文件调优，选它们没问题。两家的社区规模今天都大于 MateClaw。 + +**MateClaw 是那个给团队用的版本。** 每个 agent、每个模型、每个工具都有 RBAC。危险动作自动暂停等审批。完整审计事件流。一个 Web 管理台里，一个运维能同时管 50 个 agent 跑在 14 家供应商上。底座是 Spring Boot——任何一家已经在生产跑 Java 服务的公司可以直接并入。 + +**同一套"完整一整套"哲学，不同的重心。** --- @@ -149,16 +166,15 @@ docker compose up -d # http://localhost:18080 mateclaw/ ├── mateclaw-server/ Spring Boot 3.5 后端（Spring AI Alibaba · StateGraph 运行时） ├── mateclaw-ui/ Vue 3 + TypeScript 管理 SPA（构建产物打进后端 JAR） -├── mateclaw-desktop/ Electron 桌面端（内嵌 JRE 21） ├── mateclaw-webchat/ 网页嵌入式聊天组件（UMD / ES bundle） ├── mateclaw-plugin-api/ 第三方能力插件的 Java SDK ├── mateclaw-plugin-sample/ 参考插件实现 -├── matevip-sites/ 官网 + VitePress 文档站（pnpm workspace） -├── deploy/ 生产部署配置 ├── docker-compose.yml └── .env.example ``` +桌面端安装包通过 [GitHub Releases](https://github.com/matevip/mateclaw/releases) 分发，内嵌 JRE 21——无需安装 Java。 + ## 技术栈 | 层次 | 技术 |