claudex

MCP办公协作

提供方 Mng-dev-ai

办公协作 stdio · 本地进程 OAuth 授权 社区 / 第三方 ClaudeCursorVS CodeGemini CLI

Agentrove is a self-hosted AI coding workspace for running and orchestrating coding agents such as Antigravity, Claude Code, Codex, Copilot, Cursor, Grok, and OpenCode from one interface, combining Docker sandboxes, MCP-based multi-agent orchestration, and GitHub-assisted workflows for complex development collaboration.

一句话选型结论

按资料字段整理

claudex 属于办公协作类 MCP 服务器,stdio · 本地进程,OAuth 授权,可直接接入 Claude、Cursor、VS Code 等,没有现成安装命令,需按官方文档接入。接入前先确认权限范围;涉及写入、删除或付款的操作,建议先在测试环境验证。

分类办公协作 传输方式stdio · 本地进程 鉴权OAuth 授权 部署方式本地部署
GitHub 星标 336 仓库热度,无公开仓库时显示未标注
工具数 未标注 暴露给模型的工具数量
资源 / 提示词 0 / 0 资源与提示词模板数量
维护方 社区 / 第三方 核验等级 仅资料核对

工具与能力

办公协作
数量来自来源目录的标注,具体工具清单以官方文档为准
工具数量未标注暴露给模型可调用的工具个数
资源数量0可读取的资源条目
提示词数量0内置提示词模板数量

能力标签

Local Deployment AI coding multi-agent orchestration TypeScript self-hosted MCP server Docker sandbox GitHub integration code editor cross-platform client

支持客户端

适用场景

Local DeploymentAI codingmulti-agent orchestration

安装与配置

stdio · 本地进程 OAuth 授权
客户端在本机拉起子进程,数据不出本机
  1. 1
    准备运行环境

    stdio 服务在本地起进程,需要先装好 node,再按安装命令拉取依赖。

  2. 2
    安装服务

    复制上面的安装命令执行;命令里带占位符的位置,换成自己的密钥或路径。

  3. 3
    写入客户端配置

    在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,让新配置生效。

  4. 4
    验证是否可用

    连接成功后随便调一个只读工具,确认能返回结果,再接入正式流程。

典型使用场景

3 类
按分类与标签整理的常见用法,具体以官方文档为准
日程与消息

查日程、发消息、约会议。

表单与审批

填表单、走审批流。

团队协作

同步任务、汇总进展。

选择建议

选之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 已适配 Claude、Cursor、VS Code,换客户端不用重新研究配置。

先想清楚的地方

  • stdio 方式要求本地装好运行时(如 Node.js),换机器要重新装一遍。
  • 没有现成安装命令,需要按官方文档自己填启动命令和参数。
  • 使用前要准备好OAuth 授权对应的凭据,并按最小权限配置。
  • 资料没标注工具数量,实际能力要装好后确认。
  • 版本更新与行为细节以官方仓库为准。

更适合

  • 需要在办公协作场景里让模型直接调用外部能力的人。
  • 已经在用 Claude、Cursor 的用户,接入路径最短。
  • 想先快速试一下、能接受在本地或自有环境跑服务的人。

不太适合

  • 完全不想在本地装环境的人。
  • 不方便管理密钥或第三方授权的人。
  • 需要厂商级 SLA 和技术支持的生产环境。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
MCP 服务器工具数传输方式部署方式安装命令最后核验
claudex当前 未标注 stdio · 本地进程 本地部署 无 2026-10-08
CowAgent 未标注 stdio · 本地进程 本地部署 有 2026-04-11
Karakeep MCP 未标注 未标注 未标注 无 2026-05-17
agentscope 未标注 stdio · 本地进程 本地部署 有 2026-04-11
note-gen 未标注 stdio · 本地进程 本地部署 有 2026-04-11
Everywhere 未标注 stdio · 本地进程 本地部署 有 2026-04-12

服务器档案

办公协作
字段来自项目仓库与公开索引,中文说明由本站标注
提供方Mng-dev-ai上游标注的维护方 Mng-dev-ai
分类办公协作本站按上游分类展示
连接方式stdio · 本地进程客户端在本机拉起子进程,数据不出本机
鉴权方式OAuth 授权远程服务建议用最小权限凭据
维护方社区 / 第三方核验等级 仅资料核对
包名 / 运行时未标注 · node本地安装时用到的包标识与运行时
部署类型本地部署上游给出的部署形态
最近核验2026-10-08上游最后核验时间

详细介绍

10.0 KB
来自项目仓库与上游目录的原文,未做删改

Agentrove

Self-hosted AI coding workspace for running and orchestrating Antigravity, Claude Code, Codex, Copilot, Cursor, Grok, and OpenCode agents from one interface.

What It Does

  • Runs Antigravity, Claude, Codex, Copilot, Cursor, Grok, and OpenCode through ACP adapters.
  • Gives each workspace its own Docker or host sandbox.
  • Combines chat, code editor, terminal, file tree, diffs, secrets, and git tools in one workspace.
  • Supports workspaces from empty folders, git clones, existing local folders, or GitHub repositories.
  • Streams agent sessions with cancellation, permission prompts, queued follow-up messages, file mentions, slash commands, and attachments.
  • Includes sub-threads, pinned chats, worktree mode, personas, custom instructions, environment variables, and installed agent skills.
  • Orchestrates multi-agent workflows through the bundled MCP server: a lead chat spawns worker sub-threads on any installed agent, model, and persona — in parallel worktrees when needed — then reviews their results.
  • Hosts channels: group chats where you and several agents, each with its own model and persona, discuss and work in one shared workspace.
  • Provides GitHub-assisted repository browsing, pull request review, PR creation, reviewer selection, and git branch/commit/push/pull helpers.
  • Ships as a Docker web app, a macOS desktop app, and a native iOS app.

Orchestrating Agents

Agentrove chats aren't just endpoints — they can drive each other. The bundled MCP server (mcp-server/) exposes the whole instance as tools (send_message, get_messages, list_models, list_personas, …), and every chat's agent has those tools available. That turns any chat into an orchestrator: a lead agent on a strong model that decomposes the work, routes each task to the right agent, and reviews what comes back.

The primitives:

  • Sub-threads — send_message(parent_chat_id=…) creates a worker chat grouped under the lead chat, in the same workspace and branch. Sub-threads stay flat (no nesting): the lead fans out, workers report back.
  • Per-turn model and persona — each worker runs on any installed agent and model (model_id) with any persona (a custom system prompt). Typical fleet: a fast model with a read-only scout persona for codebase exploration, coding models for implementation, dedicated reviewer personas (bug hunting, structural quality) for QA. list_models reports each model's supported reasoning tiers (thinking_modes), so the lead dials effort per task.
  • Isolated worktrees — worktree=true gives a worker its own git worktree, so parallel workers edit concurrently without conflicts.
  • Polling and follow-ups — the lead polls get_messages until a worker's turn completes, judges the result against the actual code, and sends rework to the worker's own thread; follow-ups inherit the thread's previous model, persona, and reasoning settings.
  • Unattended turns — orchestrated turns run in the agent's full-execution mode, so workers finish without permission prompts.

A typical loop: the lead explores with a cheap fast model, hands a precise spec to a coding model in a sub-thread, fans reviewer personas out over the diff in parallel, triages their findings, delegates the accepted fixes — and owns the final result. Models, personas, and routing rules are all user-defined, so the same machinery drives whatever fleet you run.

Channels

Channels are group chats between you and multiple agents. Create one from the sidebar, pick a workspace (optionally a branch or a dedicated worktree), and add members — each with its own agent and model, display name, persona, thinking level, and permission mode.

  • Shared conversation — every member sees new channel messages in batches and either posts one short reply or passes, so agents speak only when they add something: answering, disagreeing, correcting, or proposing what others missed.
  • Shared workspace — members work in the same directory and announce which files they're about to change before editing them.
  • Live status — the status row shows which members are thinking, typing, waiting for approval, or retrying; each reply carries a collapsed "Worked" rollup with its tool activity and duration.
  • Control — attach files, approve permission prompts inline, and stop running members from the composer.

Quick Start

Requirements:

  • Docker
  • Docker Compose```bash
    git clone https://github.com/Mng-dev-ai/agentrove.git
    cd agentrove
    cp .env.example .env

Set `SECRET_KEY` in `.env`:```bash
openssl rand -hex 32

Start Agentrove:```bash
docker compose up -d


Open http://localhost:3000.

Antigravity supports authentication through Google OAuth completed in a terminal. Its ACP tokens live in `~/.gemini/antigravity-acp/`, separately from the Antigravity CLI login.

## Desktop

Agentrove also has a macOS desktop app built with Tauri. It starts a bundled Python backend sidecar on an available `127.0.0.1` port and connects the frontend to it at launch.

- Download the latest Apple Silicon build from Releases.
- Build from source:```bash
cd frontend
npm install
npm run desktop:dev

Mobile (iOS)

Agentrove also builds a native iOS app with Tauri. Since iOS can't run the local
backend sidecar, the app is a thin client: it talks to an Agentrove instance you
already host (your Docker or production deployment), reachable from the phone over
https/wss. Because the project is open source, you build and sign it yourself —
nothing is hardcoded to anyone else's server.

Requirements: macOS with Xcode (plus its iOS SDK and Simulator), the Rust iOS
targets (rustup target add aarch64-apple-ios aarch64-apple-ios-sim), and CocoaPods.

> If your device runs an iOS beta whose SDK isn't in a stable Xcode yet, build against the latest stable Xcode's SDK (sudo xcode-select -s /Applications/Xcode.app) — apps built with an older SDK still run on newer iOS.

Point the app at your instance:```bash
cd frontend
cp .env.mobile.example .env.mobile # then set your https/wss URLs
npm install


Run in the simulator:```bash
npm run ios:dev

Install on your own iPhone with a free Apple ID — no paid developer account needed
(the app must be re-signed every 7 days):

  1. Add your Apple ID in Xcode → Settings → Accounts, then under
    Manage Certificates create an Apple Development certificate.
  2. Connect your iPhone over USB, unlock it, and trust the Mac. Enable Developer
    Mode (Settings → Privacy & Security), then after the first install trust the
    certificate under Settings → General → VPN & Device Management.

Build a standalone .ipa and install it on the connected iPhone. The easiest path
is the bundled helper, which builds, signs, exports, and installs in one step:```bash
cd frontend
npm run ios:install


It signs as the team of your Apple Development certificate (override with
`APPLE_DEVELOPMENT_TEAM=`), so nothing is hardcoded to anyone else's
team. It installs on the first connected iPhone/iPad, and when that device has no
valid provisioning profile yet — a new device, or a free-team profile past its 7
days — it registers the device with your team and creates the profile first.

Or run the steps yourself (open the Xcode project once with the iPhone connected and
select it as the run destination, so Xcode registers it and creates the provisioning
profile). The `-c` flag injects the team into both the build
signing and the IPA export, so it stays out of the committed config. On Xcode 27,
`CARGO_PROFILE_RELEASE_DEBUG` works around a release-link failure
(tauri#16130):```bash
CARGO_PROFILE_RELEASE_DEBUG=line-tables-only npm run ios:build -- --export-method debugging \
  -c '{"bundle":{"iOS":{"developmentTeam":"<YOUR_TEAM_ID>"}}}'
# -> src-tauri/gen/apple/build/arm64/Agentrove.ipa

xcrun devicectl list devices   # find your device id
xcrun devicectl device install app --device <DEVICE_ID> \
  src-tauri/gen/apple/build/arm64/Agentrove.ipa

To update the app later, or to re-sign it after a free team's 7 days, re-run
npm run ios:install — the standalone build runs on the phone without keeping a Mac
connected.

Push notifications

The iOS app gets "task completed" and "permission needed" pushes from your backend
while it's in the background or the phone is locked. Tapping one opens the chat or
channel. The lock-screen text is generic; no prompts, code, or replies are sent to Apple.

Push needs a paid Apple Developer Program membership; free-team installs can't
receive it. In the developer portal:

  1. Register an App ID com.agentrove.desktop with the Push Notifications capability.
  2. Create a key with Apple Push Notifications service (APNs) enabled (Team
    Scoped, Sandbox & Production) and download the .p8.

Then set these on the backend; push stays off until all three are set:```bash
APNS_KEY="$(awk '{printf "%s\n", $0}' AuthKey_XXXXXXXXXX.p8)"
APNS_KEY_ID=XXXXXXXXXX
APNS_TEAM_ID=<YOUR_TEAM_ID>

APNS_TOPIC=com.agentrove.desktop (default; change if you use another bundle ID)


Push is opt-in on the app side because it needs the `aps-environment` entitlement,
which free teams can&#x27;t sign. Build with it enabled:```bash
cd frontend
AGENTROVE_IOS_PUSH=1 npm run ios:install

This writes the entitlement into the generated Xcode project's entitlements file (a
plain npm run ios:install removes it again). The app picks the sandbox or
production APNs host from its provisioning profile.
Pushes respect the Notifications setting. Tapping a push while the app is
suspended opens its chat, but a tap that cold-launches the app may open the app's
start page instead.

Production

For a single-host Docker deployment:bash SECRET_KEY=$(openssl rand -hex 32) \ APP_URL=https://yourdomain.com \ ALLOWED_ORIGINS=https://yourdomain.com \ docker compose -f docker-compose-production.yml up -d --build

Stack

  • Frontend: React 19, TypeScript, Vite, Tailwind CSS, Monaco, xterm.js
  • Backend: FastAPI, SQLAlchemy, SQLite, Redis
  • Runtime: ACP, Docker or host sandboxes, Tauri desktop sidecar

Community

Join the Discord server.

Contributing

Contributions and feedback are welcome.

License

Apache 2.0. See LICENSE.

常见问题

8 问
答案由本站依据条目字段整理,不引用上游原文
claudex 怎么安装?

资料没有给出安装命令,请到官方仓库或文档按说明接入;这类条目多数是生态应用,或尚未发布可安装包。

claudex 怎么在 Claude、Cursor、VS Code 里配置?

它的传输方式是stdio · 本地进程,在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,配置才会生效。

claudex 收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于服务本身和背后的接口额度,正式使用前按官方说明确认。

claudex 需要 API Key 吗?

资料标注需要OAuth 授权认证,使用前准备好对应凭据,并按最小权限配置。

claudex 支持哪些客户端?

资料标注支持 Claude、Cursor、VS Code、Gemini CLI、Codex;其他客户端只要能配置 MCP Server,也能用同样的命令或地址接入。

claudex 能提供哪些能力?

资料显示能力标签包括 Local Deployment、AI coding、multi-agent orchestration、TypeScript、self-hosted、MCP server。具体工具清单以官方文档为准。

claudex 需要什么运行环境?

运行时为 node;属于本地进程,需要在本机装好运行环境。

claudex 是本地还是远程?

它是本地进程(stdio),由客户端在本机拉起。传输方式为stdio · 本地进程。

热门专题

按标签挑服务器
全部专题 →

服务器信息以作者仓库与官方文档为准;分类、传输方式与鉴权方式的中文说明为中文整理,最近核验 2026-10-08。