@dadigua/hyperchat-core

中文 | English

HyperChat

🌟 Next Generation AI Workspace - An intelligent collaboration platform based on the workspace concept and MCP ecosystem

HyperChat 2.0 is a revolutionary multi-platform AI workspace designed with the core concept of Workspace, fully supporting the MCP (Model Context Protocol) ecosystem, allowing AI to deeply integrate with your projects to create a truly intelligent development and working environment.

🎯 Project Vision

HyperChat 1.0 is a completely manually coded project, currently migrating to version 2.0.

HyperChat 2.0 invites everyone to Vibe Coding, and encourages the use of AI tools like Claude Code and GitHub Copilot for development. Here is the project introduction:

🎯 Core Philosophy

🗂️ Workspace-Driven Intelligent Collaboration

HyperChat 2.0 centers around the Workspace, closely integrating AI capabilities with your projects:

• Project Workspace: Each project has its own independent .hyperchat/ configuration directory
• Global Workspace: System-level configuration, shareable across projects
• Workspace Isolation: Agent, MCP services, and chat logs are independently managed by workspace
• Seamless Switching: Multiple workspace tabs allow one-click switching between different project environments

🔌 Deep Integration of MCP Ecosystem

As a complete implementation of the MCP protocol, HyperChat has built a robust tools ecosystem:

• Rich Built-in MCP Services: File system, terminal, settings management, knowledge base, etc.
• Dynamic Tool Invocation: Real-time loading and management of MCP tools
• Workspace-Level Configuration: Each workspace can configure different MCP services
• Debug-Friendly: Visualized tool invocation process, supporting parameter debugging

✨ Core Features

🏢 Unified Experience Across Platforms

• 🌐 Web Application: Access directly from the browser, supporting H5 mobile
• 💻 Electron Desktop: Native application experience with full functionality support
• ⌨️ Command-Line Interface: Terminal experience similar to Claude Code
• 🔧 VSCode Plugin: IDE integrated with WebView

🤖 AI Capability Matrix

• Multi-Model Support: OpenAI, Claude, Gemini, Qwen, Deepseek, GLM, Ollama, xAI, etc.
• Intelligent Agent System: Assign preset prompts and select allowed MCP tools
• Scheduled Task Execution: Specify Agent to complete tasks on schedule
• Model Comparison Chatting: Simultaneously use multiple models for comparison testing

🎨 Modern Interface

• Tab Workspace: Intuitive management of multiple workspaces
• Smart Rendering: Supports Artifacts, SVG, HTML, Mermaid, mathematical formulas
• Dark Mode: Eye-friendly dark theme
• Multi-Language Support: Seamless switching between Chinese and English

🚀 Quick Experience 1.0

Online Demo 1.0

• HyperChat - Node.js version
• Docker Demo - ClawCloud deployment

Command Line Quick Start

npx -y @dadigua/hyper-chat

Default access: http://localhost:16100/123456/ (Password: 123456)

Docker Deployment

docker pull dadigua/hyperchat-mini:latest

🛠️ Technical Architecture

📂 Project Structure

HyperChat/
├── packages/
│   ├── core/              # Node.js core service layer
│   │   ├── src/shared/    # Shared code between front-end and back-end
│   │   ├── src/workspace/ # Workspace management system
│   │   ├── src/mcp/       # MCP service implementation
│   │   └── src/rag/       # Knowledge base RAG system
│   ├── web/               # React Web front-end
│   ├── electron/          # Electron desktop application
│   └── cli/               # Command line interface
└── docs/                  # Complete documentation

🏗️ Workspace Architecture

Project Directory/
├── .hyperchat/
│   ├── workspace.json     # Workspace configuration
│   ├── mcp.json           # MCP service configuration
│   ├── ai_models.json     # AI model configuration
│   └── agents/            # Agent configurations and chat logs
│       ├── agent-name/
│       │   ├── agent.yaml      # Agent configuration
│       │   ├── memory.md       # Agent memory
│       │   └── chatlogs/       # Chat logs
│       └── ...
└── Global Workspace: ~/Documents/HyperChat/.hyperchat/

🔄 Communication Architecture

• Unified Command Layer: packages/core/src/command.mts handles front-end and back-end communication
• AI Service Layer: packages/core/src/shared/ai.mts unifies AI model calls
• Workspace Management: packages/core/src/workspace/ provides workspace lifecycle management
• MCP Integration: packages/core/src/mcp/ implements complete MCP protocol support
• Real-Time Communication: WebSocket supports real-time message pushing and status synchronization

🎮 Feature Matrix

🗂️ Workspace Features

• ✅ Multi-Workspace Tab: Manage multiple projects within a single interface
• ✅ Workspace Isolation: Configurations, Agents, and chat logs are completely isolated
• ✅ File Tree Integration: Browse and edit project files directly in the workspace
• ✅ Workspace History: Quickly access recently used workspaces
• ✅ Global/Project Configuration: Flexible configuration inheritance mechanism

🔌 MCP Ecosystem

• ✅ Built-in MCP Services:
  • 🗃️ KnowledgeBase: Intelligent knowledge base management
  • 💾 Settings: System settings management
  • 🖥️ Terminal: Execute terminal commands
  • 🔗 Gateway: MCP service gateway
  • 🛠️ HyperTools: Super toolset
• ✅ Dynamic Service Management: Real-time start/stop of MCP services
• ✅ Tool Invocation Visualization: Clearly display AI tool usage process
• ✅ Parameter Debugging: Supports manual debugging of MCP tool parameters

🤖 AI Model Matrix

Model Provider	Recommendation	Special Features
Claude	⭐⭐⭐⭐⭐⭐	Best code understanding and generation
OpenAI	⭐⭐⭐⭐⭐	Perfect Function Calling support
Gemini Flash 2.5	⭐⭐⭐⭐⭐	High-performance inference
Qwen	⭐⭐⭐⭐	Excellent performance in Chinese scenarios
Deepseek	⭐⭐⭐⭐	Expertise in code generation
Doubao	⭐⭐⭐	Good stability

🎨 Advanced Features

• ✅ HyperPrompt Syntax: Supports variables and JS code snippets
• ✅ Intelligent Agent System: Configurable prompts and tool permissions
• ✅ Scheduled Task Execution: Supports automated workflows
• ✅ Multi-Modal Rendering: Artifacts, SVG, HTML, Mermaid, mathematical formulas
• ✅ WebDAV Sync: Cross-device configuration synchronization
• ✅ Model Comparison Chatting: Parallel testing of different models' effectiveness

📦 Environment Requirements

Make sure your system has installed:

• Node.js
• uv (recommended for Python environment management)

Install uv

# macOS
brew install uv

# Windows
winget install --id=astral-sh.uv -e

Install Node.js

# macOS
brew install node

# Windows
winget install OpenJS.NodeJS.LTS

🔧 Development Guide

💻 Local Development

# Clone the project
git clone https://github.com/BigSweetPotatoStudio/HyperChat.git
cd HyperChat

# Install dependencies
npm install
cd packages/electron && npm install
cd packages/web && npm install
cd ../..

# Start the development server
npm run dev

🌟 AI Collaborative Development

HyperChat 2.0 itself is the best practice for AI Collaborative Development:

• HyperChat 1.0 → Traditional manual development model
• HyperChat 2.0 → AI-driven Vibe Coding model

Recommended development toolchain:

• 🤖 Claude Code - Intelligent code generation and refactoring
• 🔧 GitHub Copilot - Real-time code completion
• 📝 HyperChat - AI dialogue with project context

🏗️ Workspace Development Model

• Create Project Workspace: Run HyperChat in the project directory
• Configure MCP Services: Enable file system, terminal, and other tools as needed
• Set Up Dedicated Agent: Configure specific AI assistants for the project
• AI Collaborative Coding: Utilize workspace context for intelligent development

🌟 Development Roadmap

🎯 Short-Term Goals

• Multi-Agent Collaboration: Intelligent dialogue and task division among Agents
• Deep Integration with VSCode: Complete in-IDE development experience
• MCP Plugin Store: Community discovery and installation of MCP services
• Mobile Adaptation: Responsive design optimization

🚀 Long-Term Vision

• AI-Native IDE: A fully AI and workspace-based development environment
• Team Collaboration Platform: Shared workspaces and Agents for multiple users
• Ecosystem Building: An open MCP developer community
• Intelligent Project Management: AI-driven full lifecycle project management

📸 Feature Preview

🗂️ Workspace Tab Management

• Intuitive interface for switching between multiple workspaces
• Both global workspaces and project workspaces co-exist
• Each workspace has its own file tree, Agent, and MCP configurations

🔌 MCP Service Ecosystem

• Rich built-in MCP services: knowledge base, terminal, settings management, etc.
• Visualized tool invocation process, supporting parameter debugging
• Workspace-level MCP configuration with flexible service combinations

🤖 Intelligent Agent System

• Workspace-specific Agent configurations
• Supports preset prompts and tool permissions management
• Chat logs organized by workspace and Agent

🎨 Multi-Modal Content Rendering

• HyperPrompt: Supports text + JS code variables
• Artifacts: Interactive code and interface previews
• Mathematical Formulas: KaTeX rendering support
• Chart Support: Mermaid, SVG, HTML, etc.

📊 Model Comparison Testing

• Parallel use of multiple AI models
• Real-time comparison of different models' response quality
• Supports workspace-level model configurations

🤝 Community Communication

• Telegram
• QQ Group

📄 Disclaimer

This project is for learning and communication purposes only. Any actions taken using this project (such as crawling behaviors, etc.) are not related to the project developers.

📜 License

This project is under an open-source license, please refer to the LICENSE file for details.

✨ HyperChat 2.0 - Redefining AI workspaces, making intelligent collaboration a reality ✨