cunzhi

# Cunzhi ## Say Goodbye to AI Termination Worries, Empower AI to Last Longer Cunzhi is an intelligent interaction tool based on the MCP (Model Context Protocol), providing pop-up interaction and global memory management features, making AI assistants more durable and intelligent. ## ✨ Features - 🎯 **MCP Standard Compliance**: Fully compliant with the MCP 2024-11-05 protocol specification - 🎨 **Beautiful Pop-up Interface**: A modern UI built with Vue 3 + Tailwind CSS - 🔒 **Secure Markdown Rendering**: Supports code highlighting and secure content rendering - 🖥️ **Native Application Experience**: A cross-platform desktop application based on Tauri - ⚡ **Efficient Communication**: Uses inter-process communication for quick responses - 🛠️ **Flexible Interaction**: Supports predefined options, multiple selections, and free text input - 🧠 **Global Memory Management**: Smart storage and management of development specifications, user preferences, and best practices - 📷 **Image Support**: Supports image uploads and Base64 processing ## 📸 Feature Demonstration ### Intelligent Pop-up Interface  _The modern pop-up interface of Cunzhi supports Markdown rendering, code highlighting, predefined options, and free text input._ ## 🚀 Quick Installation ### Method 1: Using Installation Script (Recommended) ```bash # Clone the repository git clone https://github.com/imhuso/cunzhi.git cd cunzhi # Run the installation script chmod +x install.sh ./install.sh ``` ### Method 2: Download Precompiled Version Download the precompiled version for your platform from the [Releases](https://github.com/imhuso/cunzhi/releases) page: - **Linux**: `cunzhi-cli-v*-linux-x86_64.tar.gz` - **macOS (Intel)**: `cunzhi-cli-v*-macos-x86_64.tar.gz` - **macOS (Apple Silicon)**: `cunzhi-cli-v*-macos-aarch64.tar.gz` - **Windows**: `cunzhi-cli-v*-windows-x86_64.zip` **Installation Steps**: 1. Download the corresponding platform's compressed package and extract it. 2. Place the `Cunzhi` and `Wait` CLI tools in any directory. 3. Optionally add them to the PATH environment variable or call them using absolute paths. **Deployment Example**: ```bash # Linux/macOS - Global Installation tar -xzf cunzhi-cli-v*-linux-x86_64.tar.gz sudo cp Cunzhi Wait /usr/local/bin/ # Linux/macOS - User Directory tar -xzf cunzhi-cli-v*-linux-x86_64.tar.gz mkdir -p ~/.local/bin cp Cunzhi Wait ~/.local/bin/ echo 'export PATH="$PATH:~/.local/bin"' >> ~/.bashrc # Windows - Custom Directory # Extract the zip file to C:\cunzhi # Optionally: Add C:\cunzhi to the system PATH ``` ## 📋 Deployment Configuration ### MCP Client Configuration Add the following configuration to your MCP client configuration file: ```json { "mcpServers": { "Cunzhi": { "command": "Cunzhi" } } } ``` **Note**: If `Cunzhi` is not in the PATH, please use the absolute path: ```json { "mcpServers": { "Cunzhi": { "command": "/path/to/Cunzhi" } } } ``` ### Tool Collaboration Mechanism Cunzhi includes two CLI tools that have an intelligent collaboration mechanism: - **Cunzhi**: MCP server that handles requests from AI clients - **Wait**: Pop-up interface that provides user interaction **Auto-discovery Logic**: 1. When the MCP server needs to display a pop-up, it will automatically look for the `Wait` tool. 2. Search order: Same directory → PATH environment variable → Error message 3. This means you can place both tools anywhere as long as they can be found. ## 🛠️ MCP Tools ### 1. zhi - Intelligent Interaction Tool A pop-up interaction tool that supports various input methods and Markdown rendering. **Main Parameters**: - `message` (required): The message content displayed to the user - `predefined_options` (optional): List of predefined options, supports multiple selections - `is_markdown` (optional): Whether to enable Markdown format rendering **Feature Highlights**: - ✅ Predefined options with multiple selections + free text input - ✅ Image uploads and Base64 processing - ✅ Markdown rendering and code highlighting - ✅ Modern dark theme UI ### 2. jiyi - Global Memory Management Tool An intelligent memory management system that stores development specifications and preferences by project. **Main Parameters**: - `action` (required): Operation type ("remember" or "recall") - `project_path` (required): Project path - `content` (required when remembering): Content to remember - `category` (optional): Memory category (rule/preference/pattern/context) **Feature Highlights**: - ✅ Automatically organizes memories by project - ✅ Intelligent category management (specifications/preferences/patterns/context) - ✅ Automatic recognition of Git repository root directory - ✅ Compressed format output to save tokens ## 📋 Usage ### Start the MCP Server ```bash Cunzhi ``` ### Start the Pop-up Interface ```bash Wait # Start the settings interface Wait --mcp-request file # MCP pop-up mode ``` ### Version Management ```bash # Check current version node scripts/version.js --current # Release a new version ./scripts/release.sh ``` ## 📁 Project Structure ``` cunzhi/ ├── src/ │ ├── mcp/tools/ │ │ ├── memory/ # Memory management tool │ │ └── interaction/ # Intelligent interaction tool │ ├── assets/sounds/ # Audio resources │ ├── main.rs # Tauri main application │ └── bin/mcp_server.rs # MCP server ├── scripts/ │ ├── release.sh # Release script │ ├── version.js # Version management │ └── README.md # Script usage instructions ├── .github/workflows/ │ └── build.yml # Cross-platform build ├── install.sh # Installation script ├── Cargo.toml # Rust configuration ├── package.json # Node.js configuration ├── tauri.conf.json # Tauri configuration └── version.json # Version configuration ``` ## 🧪 Development ### Local Development ```bash # Install dependencies pnpm install # Run frontend in development mode pnpm dev # Build and run the MCP server cargo build --release ./target/release/Cunzhi ``` ### Build Release Version ```bash # Build frontend pnpm build # Build CLI tools cargo build --release # Run installation script ./install.sh ``` ## 🔧 Troubleshooting ### Build Issues If you encounter build errors, please check the following: 1. **Linux System Dependencies**: ```bash # Ubuntu/Debian sudo apt-get install pkg-config libgtk-3-dev libasound2-dev # CentOS/RHEL sudo yum install pkgconfig gtk3-devel alsa-lib-devel ``` 2. **Windows Build**: - Ensure that Visual Studio Build Tools are installed - Use PowerShell or CMD to run the build commands 3. **macOS Build**: - Ensure that Xcode Command Line Tools are installed ```bash xcode-select --install ``` ### Runtime Issues - **Permission Issues**: Ensure that the CLI tools have execution permissions - **Path Issues**: Ensure both tools are in the same directory or both in the PATH - **Dependency Issues**: Check if the necessary runtime libraries are installed on the system ## 🤝 Contribution Contributions are welcome! Please submit Issues and Pull Requests! ## 📄 License MIT License