Skip to content
Back to search

open-cowork

Open-source AI agent desktop app for Windows & macOS. One-click install Claude Code, MCP tools, and Skills — with sandbox isolation, multi-model support, and Feishu/Slack integration.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/opencoworkai--open-cowork/badge.svg)](/listings/opencoworkai--open-cowork)

HTML

<a href="/listings/opencoworkai--open-cowork"><img src="/api/listings/opencoworkai--open-cowork/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Option 1: Homebrew (macOS, Recommended)

brew tap OpenCoworkAI/tap
brew install --cask --no-quarantine open-cowork

> The `--no-quarantine` flag bypasses macOS Gatekeeper, so you won't see the "Apple cannot verify this app" warning.

Option 2: Download Installer

Get the latest version from our [Releases Page](https://github.com/OpenCoworkAI/open-cowork/releases).

| Platform | File Type | | ------------------------- | --------- | | **Windows** | `.exe` | | **macOS** (Apple Silicon) | `.dmg` |

Option 3: Build from Source

For developers who want to contribute or modify the codebase:

git clone https://github.com/OpenCoworkAI/open-cowork.git
cd open-cowork
npm install
npm run rebuild
npm run dev

To build the installer locally: `npm run build`

Security Configuration: 🔒 Sandbox Support

Open Cowork provides **multi-level sandbox protection** to keep your system safe:

| Level | Platform | Technology | Description | | ------------ | -------- | ---------- | ---------------------------------------------- | | **Basic** | All | Path Guard | File operations restricted to workspace folder | | **Enhanced** | Windows | WSL2 | Commands execute in isolated Linux VM | | **Enhanced** | macOS | Lima | Commands execute in isolated Linux VM |

- **Windows (WSL2)**: When WSL2 is detected, all Bash commands are automatically routed to a Linux VM. The workspace is synced bidirectionally. - **macOS (Lima)**: When [Lima](https://lima-vm.io/) is installed (`brew install lima`), commands run in an Ubuntu VM with `/Users` mounted. - **Fallback**: If no VM is available, commands run natively with path-based restrictions.

**Setup (Optional, Recommended)**

- **Windows**: WSL2 is auto-detected if installed. [Install WSL2](https://docs.microsoft.com/en-us/windows/wsl/install)

- **macOS**: Lima is auto-detected if installed. Install command:

brew install lima
# Open Cowork will automatically create and manage a Lima VM (internal Lima name: 'claude-sandbox')

---

<a id="quick-start"></a>

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

readme.md · README · 352 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <p align="center">
  2. <img src="resources/logo.png" alt="Open Cowork Logo" width="280" />
  3. </p>
  4. <h1 align="center">🚀 Open Cowork: Your Personal AI Agent Desktop App</h1>
  5. <p align="center">
  6. • Open Source Claude Cowork • One-Click Install
  7. </p>
  8. <p align="center">
  9. <a href="./README_zh.md">中文文档</a> •
  10. <a href="#features">Features</a> •
  11. <a href="#demo">Demo</a> •
  12. <a href="#installation">Downloads</a> •
  13. <a href="#quick-start">Quick Start</a> •
  14. <a href="#skills">Skills Library</a>
  15. </p>
  16. <p align="center">
  17. <img src="https://img.shields.io/badge/Platform-Windows%20%7C%20macOS-blue" alt="Platform" />
  18. <img src="https://img.shields.io/badge/License-MIT-green" alt="License" />
  19. <img src="https://img.shields.io/badge/Node.js-18+-brightgreen" alt="Node.js" />
  20. <a href="https://discord.gg/pynjtQDf"><img src="https://img.shields.io/discord/1493588403260883078?logo=discord&label=Discord&color=5865F2" alt="Discord" /></a>
  21. <a href="#community"><img src="https://img.shields.io/badge/WeChat-微信群-07C160?logo=wechat&logoColor=white" alt="WeChat" /></a>
  22. </p>
  23. ---
  24. Open Cowork is a free, open-source AI agent desktop application for Windows and macOS. It wraps Claude Code, OpenAI, Gemini, DeepSeek, and other AI models into a user-friendly GUI with one-click installation — no coding required. Key capabilities include VM-level sandbox isolation (WSL2 on Windows, Lima on macOS), a built-in Skills system for generating PPTX, DOCX, XLSX, and PDF documents, MCP (Model Context Protocol) integration for connecting to browsers, Notion, and other desktop apps, GUI automation via computer use, and remote control through Feishu (Lark) and Slack. Open Cowork is the open-source implementation of Claude Cowork, designed to make AI-powered desktop automation accessible to everyone.
  25. ---
  26. ## 📖 Introduction
  27. **Open Cowork** is an open-source implementation of **Claude Cowork**, with one-click installers for **Windows** and **macOS**—no coding required.
  28. It provides a sandboxed workspace where AI can manage files, generate professional outputs (PPTX, DOCX, XLSX, etc.) through our built-in **Skills** system, and **connect to desktop apps via MCP** (browser, Notion, etc.) for better collaboration.
  29. > [!WARNING]
  30. > **Disclaimer**: Open Cowork is an AI collaboration tool. Please exercise caution with its operations, especially when authorizing file modifications or deletions. We support VM-based sandbox isolation, but some operations may still carry risks.
  31. ---
  32. <a id="features"></a>
  33. ## ✨ Key Features
  34. | | MCP & Skills | Remote Control | GUI Operation |
  35. | ------------- | ------------ | -------------- | ------------- |
  36. | Claude Cowork | ✓ | ✗ | ✗ |
  37. | OpenClaw | ✓ | ✓ | ✗ |
  38. | OpenCowork | ✓ | ✓ | ✓ |
  39. - **One-Click Install, Ready to Use**: Pre-built installers for Windows and macOS, no environment setup needed—just download and start using.
  40. - **Flexible Model Support**: Supports **Claude**, **OpenAI-compatible APIs**, and Chinese models like **GLM**, **MiniMax**, **Kimi**. Use your OpenRouter, Anthropic, or other API keys with flexible configuration. More models coming soon!
  41. - **Remote Control**: Connect to collaboration platforms like **Feishu (Lark)** and other remote services to automate workflows and cross-platform operations.
  42. - **GUI Operation**: Control and interact with various desktop GUI applications on your computer. **Recommended model: Gemini-3-Pro** for optimal GUI understanding and control.
  43. - **Smart File Management**: Read, write, and organize files within your workspace.
  44. - **Skills System**: Built-in workflows for PPTX, DOCX, PDF, XLSX generation and processing. **Supports custom skill creation and deletion.**
  45. - **MCP External Service Support**: Integrate browser, Notion, custom apps and more through **MCP Connectors** to extend AI capabilities.
  46. - **Multimodal Input**: Drag & drop files and images directly into the chat input for seamless multimodal interaction.
  47. - **Real-time Trace**: Watch AI reasoning and tool execution in the Trace Panel.
  48. - **Secure Workspace**: All operations confined to your chosen workspace folder.
  49. - **VM-Level Isolation**: WSL2 (Windows) and Lima (macOS) VM isolation—all commands execute in an isolated VM to protect your host system.
  50. - **UI Enhancements**: Beautiful and flexible UI design, system language switching, comprehensive MCP/Skills/Tools call display.
  51. <a id="demo"></a>
  52. ## 🎬 Demo
  53. See Open Cowork in action:
  54. ### 1. Folder Organization & Cleanup 📂
  55. https://github.com/user-attachments/assets/dbeb0337-2d19-4b5d-a438-5220f2a87ca7
  56. ### 2. Generate PPT from Files 📊
  57. https://github.com/user-attachments/assets/30299ded-0260-468f-b11d-d282bb9c97f2
  58. ### 3. Generate XLSX Spreadsheets 📉
  59. https://github.com/user-attachments/assets/f57b9106-4b2c-4747-aecd-a07f78af5dfc
  60. ### 4. GUI Operation🖥
  61. https://github.com/user-attachments/assets/75542c76-210f-414d-8182-1da988c148f2
  62. ### 5. Remote control with Feishu(Lark) 🤖
  63. https://github.com/user-attachments/assets/05a703de-c0f5-407b-9a43-18b6a172fd74
  64. ---
  65. <a id="installation"></a>
  66. ## 📦 Installation
  67. ### Option 1: Homebrew (macOS, Recommended)
  68. ```bash
  69. brew tap OpenCoworkAI/tap
  70. brew install --cask --no-quarantine open-cowork
  71. ```
  72. > The `--no-quarantine` flag bypasses macOS Gatekeeper, so you won't see the "Apple cannot verify this app" warning.
  73. ### Option 2: Download Installer
  74. Get the latest version from our [Releases Page](https://github.com/OpenCoworkAI/open-cowork/releases).
  75. | Platform | File Type |
  76. | ------------------------- | --------- |
  77. | **Windows** | `.exe` |
  78. | **macOS** (Apple Silicon) | `.dmg` |
  79. ### Option 3: Build from Source
  80. For developers who want to contribute or modify the codebase:
  81. ```bash
  82. git clone https://github.com/OpenCoworkAI/open-cowork.git
  83. cd open-cowork
  84. npm install
  85. npm run rebuild
  86. npm run dev
  87. ```
  88. To build the installer locally: `npm run build`
  89. ### Security Configuration: 🔒 Sandbox Support
  90. Open Cowork provides **multi-level sandbox protection** to keep your system safe:
  91. | Level | Platform | Technology | Description |
  92. | ------------ | -------- | ---------- | ---------------------------------------------- |
  93. | **Basic** | All | Path Guard | File operations restricted to workspace folder |
  94. | **Enhanced** | Windows | WSL2 | Commands execute in isolated Linux VM |
  95. | **Enhanced** | macOS | Lima | Commands execute in isolated Linux VM |
  96. - **Windows (WSL2)**: When WSL2 is detected, all Bash commands are automatically routed to a Linux VM. The workspace is synced bidirectionally.
  97. - **macOS (Lima)**: When [Lima](https://lima-vm.io/) is installed (`brew install lima`), commands run in an Ubuntu VM with `/Users` mounted.
  98. - **Fallback**: If no VM is available, commands run natively with path-based restrictions.
  99. **Setup (Optional, Recommended)**
  100. - **Windows**: WSL2 is auto-detected if installed. [Install WSL2](https://docs.microsoft.com/en-us/windows/wsl/install)
  101. - **macOS**:
  102. Lima is auto-detected if installed. Install command:
  103. ```bash
  104. brew install lima
  105. # Open Cowork will automatically create and manage a Lima VM (internal Lima name: 'claude-sandbox')
  106. ```
  107. ---
  108. <a id="quick-start"></a>
  109. ## 🚀 Quick Start Guide
  110. ### 1. Get an API Key
  111. You need an API key to power the agent. We support **OpenRouter**, **Anthropic**, and various cost-effective **Chinese Models**.
  112. | Provider | Get Key / Coding Plan | Base URL (Required) | Recommended Model |
  113. | ------------------ | -------------------------------------------------------------------------- | ---------------------------------------- | -------------------- |
  114. | **OpenRouter** | [OpenRouter](https://openrouter.ai/) | `https://openrouter.ai/api` | `claude-4-5-sonnet` |
  115. | **Anthropic** | [Anthropic Console](https://console.anthropic.com/) | (Default) | `claude-4-5-sonnet` |
  116. | **Zhipu AI (GLM)** | [GLM Coding Plan](https://bigmodel.cn/glm-coding) (⚡️Chinese Deal) | `https://open.bigmodel.cn/api/anthropic` | `glm-4.7`, `glm-4.6` |
  117. | **MiniMax** | [MiniMax Coding Plan](https://platform.minimaxi.com/subscribe/coding-plan) | `https://api.minimaxi.com/anthropic` | `minimax-m2` |
  118. | **Kimi** | [Kimi Coding Plan](https://www.kimi.com/membership/pricing) | `https://api.kimi.com/coding/` | `kimi-k2` |
  119. ### 2. Configure
  120. 1. Open the app and click the ⚙️ **Settings** icon in the bottom left.
  121. 2. Paste your **API Key**.
  122. 3. **Crucial**: Set the **Base URL** according to the table above (especially for Zhipu/MiniMax, etc.).
  123. 4. Enter the **Model** name you want to use.
  124. ### 3. Start Coworking
  125. 1. **Select a Workspace**: Choose a folder where Claude is allowed to work.
  126. 2. **Enter a Prompt**:
  127. > "Read the financial_report.csv in this folder and create a PowerPoint summary with 5 slides."
  128. ### 📝 Important Notes
  129. 1. **macOS Installation**: If you downloaded the DMG directly (not via Homebrew) and see a security warning, go to **System Settings > Privacy & Security** and click **Open Anyway**. Or install via Homebrew to avoid this entirely:
  130. ```bash
  131. brew tap OpenCoworkAI/tap && brew install --cask --no-quarantine open-cowork
  132. ```
  133. 2. **Network Access**: For tools like `WebSearch`, you may need to enable "Virtual Network Interface" (TUN Mode) in your proxy settings to ensure connectivity.
  134. 3. **Notion Connector**: Besides setting the integration token, you also need to add connections in a root page. See https://www.notion.com/help/add-and-manage-connections-with-the-api for more details.
  135. ---
  136. <a id="skills"></a>
  137. ## 🧰 Skills Library
  138. Open Cowork ships with built-in skills under `.claude/skills/`, and supports user-added or custom skills, including:
  139. - `pptx` for PowerPoint generation
  140. - `docx` for Word document processing
  141. - `pdf` for PDF handling and forms
  142. - `xlsx` for Excel spreadsheet support
  143. - `skill-creator` for creating custom skills
  144. ---
  145. ## 🏗️ Architecture
  146. ```
  147. open-cowork/
  148. ├── src/
  149. │ ├── main/ # Electron Main Process (Node.js)
  150. │ │ ├── index.ts # Main entry point
  151. │ │ ├── claude/ # Agent SDK & Runner
  152. │ │ │ └── agent-runner.ts # AI agent execution logic
  153. │ │ ├── config/ # Configuration management
  154. │ │ │ └── config-store.ts # Persistent settings storage
  155. │ │ ├── db/ # Database layer
  156. │ │ │ └── database.ts # SQLite/data persistence
  157. │ │ ├── ipc/ # IPC handlers
  158. │ │ ├── memory/ # Memory management
  159. │ │ │ └── memory-manager.ts
  160. │ │ ├── sandbox/ # Security & Path Resolution
  161. │ │ │ └── path-resolver.ts # Sandboxed file access
  162. │ │ ├── session/ # Session management
  163. │ │ │ └── session-manager.ts
  164. │ │ ├── skills/ # Skill Loader & Manager
  165. │ │ │ └── skills-manager.ts
  166. │ │ └── tools/ # Tool execution
  167. │ │ └── tool-executor.ts # Tool call handling
  168. │ ├── preload/ # Electron preload scripts
  169. │ │ └── index.ts # Context bridge setup
  170. │ └── renderer/ # Frontend UI (React + Tailwind)
  171. │ ├── App.tsx # Root component
  172. │ ├── main.tsx # React entry point
  173. │ ├── components/ # UI Components
  174. │ │ ├── ChatView.tsx # Main chat interface
  175. │ │ ├── ConfigModal.tsx # Settings dialog
  176. │ │ ├── ContextPanel.tsx # File context display
  177. │ │ ├── MessageCard.tsx # Chat message component
  178. │ │ ├── PermissionDialog.tsx
  179. │ │ ├── Sidebar.tsx # Navigation sidebar
  180. │ │ ├── Titlebar.tsx # Custom window titlebar
  181. │ │ ├── TracePanel.tsx # AI reasoning trace
  182. │ │ └── WelcomeView.tsx # Onboarding screen
  183. │ ├── hooks/ # Custom React hooks
  184. │ │ └── useIPC.ts # IPC communication hook
  185. │ ├── store/ # State management
  186. │ │ └── index.ts
  187. │ ├── styles/ # CSS styles
  188. │ │ └── globals.css
  189. │ ├── types/ # TypeScript types
  190. │ │ └── index.ts
  191. │ └── utils/ # Utility functions
  192. ├── .claude/
  193. │ └── skills/ # Default Skill Definitions
  194. │ ├── pptx/ # PowerPoint generation
  195. │ ├── docx/ # Word document processing
  196. │ ├── pdf/ # PDF handling & forms
  197. │ ├── xlsx/ # Excel spreadsheet support
  198. │ └── skill-creator/ # Skill development toolkit
  199. ├── resources/ # Static Assets (icons, images)
  200. ├── electron-builder.yml # Build configuration
  201. ├── vite.config.ts # Vite bundler config
  202. └── package.json # Dependencies & scripts
  203. ```
  204. ---
  205. ## 🗺️ Roadmap
  206. See our full **[ROADMAP.md](ROADMAP.md)** for detailed plans.
  207. **Completed:** Core installers · Filesystem sandboxing · VM isolation (WSL2/Lima) · Skills (PPTX/DOCX/PDF/XLSX) · MCP connectors · Multi-model support · Rich input · i18n · GUI operation
  208. **Coming next:** Memory optimization · Linux support · Plugin system · Stable release
  209. ---
  210. ## ❓ FAQ
  211. **What is Open Cowork?**
  212. Open Cowork is a free, open-source desktop application that provides a local AI agent workspace. It wraps AI models (Claude, GPT, Gemini, DeepSeek, etc.) into a GUI with one-click installers for Windows and macOS — no terminal or coding knowledge required.
  213. **How is Open Cowork different from Claude Cowork?**
  214. Open Cowork is the open-source implementation of Claude Cowork. It adds multi-model support (not just Claude), GUI automation via computer use, remote control through Feishu/Slack, and VM-level sandbox isolation. See the [feature comparison table](#features) for details.
  215. **What AI models does Open Cowork support?**
  216. Claude (via Anthropic or OpenRouter), OpenAI-compatible APIs, and Chinese models including GLM (Zhipu AI), MiniMax, and Kimi. Any provider offering an OpenAI-compatible API endpoint can be configured.
  217. **Is Open Cowork free?**
  218. Yes. Open Cowork itself is completely free and open-source under the MIT license. You only need to pay for the AI model API usage from your chosen provider.
  219. **Does Open Cowork work on Linux?**
  220. Currently, Open Cowork provides pre-built installers for Windows and macOS only. Linux users can build from source — see the [Build from Source](#installation) section.
  221. **How does sandbox isolation work?**
  222. Open Cowork offers multi-level protection: basic path-based restrictions on all platforms, and enhanced VM-level isolation using WSL2 (Windows) or Lima (macOS). When a VM is available, all commands execute inside an isolated Linux environment, protecting your host system.
  223. **What are Skills and how do I create custom ones?**
  224. Skills are built-in workflows for specific tasks like generating PPTX, DOCX, PDF, or XLSX files. Open Cowork ships with default skills under `.claude/skills/` and includes a `skill-creator` tool to help you build your own custom skills.
  225. **What is MCP and how does it work?**
  226. MCP (Model Context Protocol) lets AI connect to external tools and services. Open Cowork supports MCP connectors for browsers, Notion, and other desktop apps — extending the AI's capabilities beyond just file management and code.
  227. **How do I set up remote control via Feishu or Slack?**
  228. Open Cowork supports remote control through Feishu (Lark) and Slack integration, allowing you to send commands and receive results from collaboration platforms. Check the app settings for remote control configuration.
  229. **Is my data safe? Does Open Cowork send data to external servers?**
  230. Open Cowork runs locally on your machine. Your files stay in your workspace. The only external communication is with the AI model API you configure (e.g., Anthropic, OpenRouter). No data is sent to Open Cowork servers.
  231. ---
  232. ## 🛠️ Contributing
  233. We welcome contributions! Whether it's a new Skill, a UI fix, or a security improvement:
  234. 1. Fork the repo.
  235. 2. Create a branch (`git checkout -b feature/NewSkill`).
  236. 3. Submit a PR.
  237. ---
  238. ## 💬 Community
  239. Join our community for support and discussion:
  240. - **Discord**: [Join our Discord server](https://discord.gg/pynjtQDf) — for real-time chat, support, and development discussion.
  241. - **WeChat**: Scan the QR code below to join our WeChat group (Chinese community).
  242. <p align="center">
  243. <img src="resources/WeChat.jpg" alt="WeChat Group" width="200" />
  244. </p>
  245. ---
  246. ## 📄 License
  247. MIT © Open Cowork Team
  248. ---
  249. <p align="center">
  250. Made with ❤️ by the Open Cowork Team with the help of opus4.5
  251. </p>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.