Open-Claude-Cowork  by caiqinghua

Desktop AI collaborator for code and non-technical tasks

Created 2 months ago
258 stars

Top 98.0% on SourcePulse

GitHubView on GitHub
Project Summary

Open Claude Cowork provides a desktop AI assistant designed to enhance productivity for programming, file management, and general tasks by offering a visual interface for AI collaboration. It targets users who find terminal-based AI tools like Claude Code limiting, providing a more interactive and manageable experience. The primary benefit is enabling users to leverage various Anthropic-compatible LLMs for complex tasks without requiring a Claude Max subscription, all within a native desktop application.

How It Works

This project is a native desktop application built using Electron, with a frontend developed in React and styled with Tailwind CSS. It acts as a graphical user interface (GUI) wrapper for the Anthropic Agent SDK, allowing users to define and execute tasks. The core advantage lies in its visual feedback mechanisms, including token-by-token streaming output, Markdown rendering with syntax highlighting, and visualized tool calls, which are absent in terminal-only interfaces. It also offers robust session management and explicit tool permission controls for enhanced user oversight.

Quick Start & Requirements

  • Installation: Download a pre-built release from the Releases page or build from source.
  • Prerequisites: Bun or Node.js 22+, Claude Code installed and authenticated.
  • Build from Source:
    1. Clone the repository: git clone https://github.com/caiqinghua/open-claude-cowork.git
    2. Navigate to the directory: cd open-claude-cowork
    3. Install dependencies: bun install
    4. Configure environment: cp .env.example .env
    5. Run in development: bun run dev
    6. Build production binaries: bun run dist:mac (macOS), bun run dist:win (Windows), bun run dist:linux (Linux).
  • Documentation: Links to Releases and the GitHub repository are provided.

Highlighted Details

  • AI Collaboration Partner: Capable of writing/editing code, managing files, creating presentations, performing topic analysis, and writing articles.
  • Session Management: Supports custom working directories, resuming conversations, local history storage (SQLite), safe deletion, and automatic persistence.
  • Real-Time Streaming Output: Features token-by-token streaming, visualization of Claude's reasoning process, and rendered Markdown with syntax-highlighted code.
  • Tool Permission Control: Requires explicit user approval for sensitive actions via interactive decision panels.
  • Claude Code Compatibility: Reuses existing ~/.claude/settings.json for seamless integration with Claude Code configurations, including API keys, base URLs, and models.
  • Model Support: Compatible with Zhipu GLM 4.7, MiniMax 2.1, Moonshot Kimi, DeepSeek, and any other Anthropic-compatible API model.

Maintenance & Community

The project welcomes contributions via pull requests. Specific community channels (like Discord/Slack) or details on notable contributors, sponsorships, or partnerships are not detailed in the README. A roadmap indicates planned features such as GUI-based configuration for models and API keys.

Licensing & Compatibility

The project is released under the MIT License, which generally permits commercial use and integration into closed-source projects.

Limitations & Caveats

Users may encounter issues with native modules like better-sqlite3 requiring recompilation specifically for Electron's bundled Node.js version using electron-rebuild. The project appears to be under active development, with a roadmap suggesting ongoing feature additions and refinements.

Health Check
Last Commit

2 months ago

Responsiveness

Inactive

Pull Requests (30d)
1
Issues (30d)
2
Star History
48 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.