AI导哈

多智能体AI开发平台

AURA AI Copilot 是一款企业级AI开发平台，通过多智能体编排、RAG知识库和全栈技术能力，帮助开发团队高效完成代码生成、项目管理、测试部署等复杂开发任务。

  • by muffy86
  • TypeScript
  • 本地部署
  • 更新于
在 GitHub 查看 muffy86/aura-ai-copilot

使用说明

AURA AI Copilot

Autonomous Unified Reality Architect - An enterprise-grade AI development platform with multi-agent orchestration, RAG, and full-stack capabilities.

🚀 Features

Multi-Agent AI System

AURA coordinates a team of specialized AI agents to tackle complex development tasks:

  • 🎯 OrchestratorAgent - Lead architect that coordinates all agents
  • 💻 CodeAgent - Code generation, refactoring, and debugging
  • 🔍 ResearchAgent - Information gathering and analysis
  • 🎨 UIAgent - UI/UX design and component creation
  • ✅ QAAgent - Testing and quality assurance
  • 🚀 DevOpsAgent - Deployment and infrastructure
  • 📊 BizOpsAgent - Project management and business operations

Core Capabilities

  • Intelligent Chat Interface - Natural language conversations with context-aware agent routing
  • Project Management - Create and manage AI-powered development projects
  • Knowledge Base - RAG-powered document storage with semantic search
  • Task Tracking - Monitor agent activities and task execution
  • Real-time Streaming - Live AI responses with markdown rendering
  • Dark/Light Theme - Beautiful purple-themed UI with theme switching

🛠️ Tech Stack

Frontend

  • React 19 with TypeScript
  • Tailwind CSS 4 for styling
  • tRPC for type-safe API calls
  • Wouter for routing
  • Shadcn/ui components
  • Streamdown for markdown rendering

Backend

  • Node.js with Express
  • tRPC 11 for API layer
  • Drizzle ORM for database
  • MySQL/TiDB database
  • OpenAI API for LLM integration

Infrastructure

  • Manus OAuth for authentication
  • S3 for file storage
  • Vector embeddings for RAG (planned)
  • MCP protocol support (planned)
  • GitHub Actions CI (type-check + build)

📦 Installation

Prerequisites

  • Node.js 22+
  • pnpm
  • MySQL or TiDB database

Setup

  1. Clone the repository:
git clone https://github.com/muffy86/aura-ai-copilot.git
cd aura-ai-copilot
  1. Install dependencies:
pnpm install
  1. Set up environment variables (managed by Manus platform):
  • DATABASE_URL - MySQL connection string
  • JWT_SECRET - Session signing secret
  • BUILT_IN_FORGE_API_KEY - LLM API key
  • Other platform-provided variables
  1. Push database schema:
pnpm db:push
  1. Start development server:
pnpm dev

The application will be available at http://localhost:3000

🎯 Usage

Starting a Conversation

  1. Navigate to the Dashboard
  2. Click New Chat
  3. Select an AI agent (or use Orchestrator for automatic routing)
  4. Start asking questions or requesting assistance

Creating a Project

  1. Go to Projects page
  2. Click New Project
  3. Enter project name and description
  4. The AI agents can now help with project-specific tasks

Adding Knowledge

  1. Visit Knowledge Base page
  2. Click Add Document
  3. Paste documentation, code, or any reference material
  4. AI agents will use this context in their responses

🏗️ Architecture

Database Schema

  • users - Authentication and user management
  • projects - User projects with specifications
  • conversations - Chat conversations
  • messages - Individual messages with agent tracking
  • tasks - Agent task execution tracking
  • knowledgeBase - RAG document storage
  • toolCalls - MCP tool usage logging

Agent System

Each agent has a specialized system prompt and expertise:

type AgentType = 
  | "orchestrator"  // Coordinates all agents
  | "code"          // Code generation
  | "research"      // Information gathering
  | "ui"            // UI/UX design
  | "qa"            // Testing
  | "devops"        // Deployment
  | "bizops";       // Project management

API Structure

All API endpoints are type-safe tRPC procedures:

  • projects.* - Project CRUD operations
  • conversations.* - Conversation management
  • chat.send - Send messages to AI agents
  • tasks.* - Task tracking
  • knowledge.* - Knowledge base management

🔮 Roadmap

Phase 1: RAG System ✨

  • [ ] Vector database integration
  • [ ] Document chunking and embedding
  • [ ] Semantic search functionality
  • [ ] Context retrieval for agents

Phase 2: MCP Protocol 🔗

  • [ ] Model Context Protocol implementation
  • [ ] Tool registry system
  • [ ] Context passing between agents
  • [ ] Tool call tracking

Phase 3: Advanced Features 🚀

  • [ ] Voice transcription support
  • [ ] Image generation for UI mockups
  • [ ] Collaborative editing
  • [ ] Workflow automation
  • [ ] Analytics and monitoring

📝 Development

Project Structure

aura-ai-copilot/
├── client/              # Frontend React application
│   ├── src/
│   │   ├── pages/      # Page components
│   │   ├── components/ # Reusable UI components
│   │   └── lib/        # tRPC client setup
├── server/              # Backend Node.js server
│   ├── agents.ts       # AI agent logic
│   ├── db.ts           # Database queries
│   └── routers.ts      # tRPC procedures
├── drizzle/            # Database schema
└── shared/             # Shared types and constants

Key Commands

  • pnpm dev - Start development server
  • pnpm build - Build for production
  • pnpm db:push - Push database schema changes
  • pnpm type-check - Run TypeScript checks

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

MIT License - see LICENSE file for details

🙏 Acknowledgments

  • Built with Manus platform
  • Powered by OpenAI's language models
  • UI components from shadcn/ui

📧 Contact

For questions or support, please open an issue on GitHub.

AURA AI Copilot - Empowering developers with AI-powered multi-agent assistance 🚀

相关 MCP