内容
# Letta Chat Assistant
A modern, professional Next.js chat application powered by [Letta](https://docs.letta.com/), featuring stateful AI agents with persistent memory and an advanced Claude-style interface.
  
## ✨ Features
### 🧠 **Stateful AI Capabilities**
- **Persistent Memory**: Agent remembers conversations across sessions
- **Learning & Adaptation**: Improves responses based on interaction history
- **Built-in Tools**: Web search and code execution capabilities
- **Memory Blocks**: Custom persona, human context, and project awareness
### 🎨 **Modern User Interface**
- **Claude-Style Layout**: Professional sidebar with chat history
- **Real-time Streaming**: Live response streaming with typing indicators
- **Markdown & Code Support**: Full syntax highlighting for 100+ languages
- **Dark Mode**: Automatic theme switching with system preferences
- **Responsive Design**: Works perfectly on desktop, tablet, and mobile
### 🚀 **Advanced Chat Features**
- **New Chat Sessions**: Start fresh conversations with one click
- **Message Actions**: Copy any message (user or assistant) with hover controls
- **Auto-scroll**: Smooth scrolling to new messages
- **Error Handling**: Graceful error display and recovery
- **Suggestion Buttons**: Quick-start prompts for new conversations
### 🎯 **Developer Experience**
- **TypeScript**: Full type safety throughout the application
- **shadcn/ui**: Beautiful, accessible UI components
- **Tailwind CSS**: Utility-first styling with custom design system
- **Hot Reload**: Instant development feedback
- **Clean Architecture**: Well-organized component structure
## 🚀 Quick Start
### Prerequisites
- **Node.js 18+** installed
- **Letta Cloud Account** - Get one free at [app.letta.com](https://app.letta.com)
### 1. Clone & Install
```bash
git clone https://github.com/jyotirmi/letta-chatbot-iqagent.git
cd letta-chatbot-iqagent
npm install
```
### 2. Get Your Letta API Key
1. Sign up at [https://app.letta.com](https://app.letta.com)
2. Navigate to [API Keys](https://app.letta.com/api-keys)
3. Create a new API key and copy it
### 3. Configure Environment
```bash
# Copy environment template
cp env.example .env.local
# Edit .env.local and add your API key:
# LETTA_API_KEY=your_api_key_here
```
### 4. Create Your Agent
```bash
# This creates a new Letta agent with custom memory blocks
npm run setup-agent
```
The setup script will:
- Create an agent named "Alex"
- Configure memory blocks (persona, human, project context)
- Enable built-in tools (web_search, run_code)
- Save the agent ID to your environment
### 5. Start the Application
```bash
npm run dev
```
Open [http://localhost:3000](http://localhost:3000) and start chatting! 🎉
## 🏗️ Project Structure
```
├── app/ # Next.js App Router
│ ├── api/chat/route.ts # Streaming chat API endpoint
│ ├── globals.css # Global styles & CSS variables
│ ├── layout.tsx # Root layout with metadata
│ └── page.tsx # Main application page
├── components/
│ └── ui/ # UI Components
│ ├── main-chat.tsx # Main chat interface with input
│ ├── enhanced-chat.tsx # Message list with auto-scroll
│ ├── sidebar.tsx # Navigation sidebar
│ ├── message.tsx # Individual message component
│ ├── suggestion-buttons.tsx # Quick-start prompts
│ ├── button.tsx # Reusable button component
│ ├── textarea.tsx # Auto-resizing text input
│ ├── avatar.tsx # User/AI avatars
│ └── scroll-area.tsx # Custom scroll container
├── lib/
│ └── utils.ts # Utility functions (cn helper)
├── scripts/
│ └── setup-agent.ts # Agent creation & configuration
└── env.example # Environment variables template
```
## 🔧 Key Technologies
| Technology | Purpose | Version |
|------------|---------|---------|
| **[Letta](https://docs.letta.com/)** | Stateful AI agent platform | Latest |
| **[Next.js](https://nextjs.org/)** | React framework with App Router | 15.x |
| **[Vercel AI SDK](https://sdk.vercel.ai/)** | Streaming AI responses | 4.x |
| **[shadcn/ui](https://ui.shadcn.com/)** | Beautiful UI components | Latest |
| **[Tailwind CSS](https://tailwindcss.com/)** | Utility-first CSS framework | 3.x |
| **[React Markdown](https://github.com/remarkjs/react-markdown)** | Markdown rendering | 10.x |
| **[Syntax Highlighter](https://github.com/react-syntax-highlighter/react-syntax-highlighter)** | Code syntax highlighting | 15.x |
| **[Lucide React](https://lucide.dev/)** | Beautiful icons | Latest |
## 💡 How It Works
### Architecture Overview
```mermaid
graph TD
A[User Input] --> B[Next.js API Route]
B --> C[Vercel AI SDK]
C --> D[Letta Agent]
D --> E[Memory Blocks]
D --> F[Built-in Tools]
F --> G[Web Search]
F --> H[Code Execution]
D --> I[Streaming Response]
I --> J[React UI]
J --> K[Markdown Rendering]
J --> L[Syntax Highlighting]
```
### Stateful vs Stateless
| Feature | Traditional Chatbots | Letta Agents |
|---------|---------------------|--------------|
| **Memory** | No persistent memory | Full conversation history |
| **Context** | Limited to current session | Cross-session context retention |
| **Learning** | Static responses | Adaptive, learning responses |
| **Tools** | Limited or no tools | Built-in web search & code execution |
| **Personalization** | Generic responses | Personalized based on user history |
### Message Flow
1. **User Input** → Captured in textarea with auto-resize
2. **API Processing** → Next.js route validates and forwards to Letta
3. **Agent Processing** → Letta agent accesses full conversation history
4. **Tool Execution** → Agent can search web or execute code if needed
5. **Response Generation** → Streaming response with markdown support
6. **UI Rendering** → Real-time updates with syntax highlighting
## 🎨 UI Components
### Chat Interface
- **Sidebar**: Navigation with chat history and new chat button
- **Main Chat**: Message list with auto-scroll and typing indicators
- **Input Area**: Auto-resizing textarea with send button and attachment placeholder
- **Message Bubbles**: Distinct styling for user (gray) and assistant (white) messages
### Message Features
- **Markdown Support**: Headers, lists, links, tables, blockquotes
- **Code Highlighting**: 100+ programming languages with VS Code Dark+ theme
- **Copy Functionality**: One-click copy for any message
- **Action Buttons**: Copy, like, dislike for assistant messages
- **Timestamps**: Formatted time display
### Responsive Design
- **Desktop**: Full sidebar with spacious chat area
- **Tablet**: Collapsible sidebar with touch-friendly controls
- **Mobile**: Stack layout with optimized input handling
## 🛠️ Customization
### Modify Agent Personality
Edit `scripts/setup-agent.ts`:
```typescript
const persona = {
name: "persona",
value: "Your custom personality here...",
description: "The agent's core personality and behavior"
}
```
### Add Custom Tools
```typescript
// Enable additional tools during agent creation
const tools = ["web_search", "run_code", "your_custom_tool"]
```
### Customize UI Theme
Edit `app/globals.css` CSS variables:
```css
:root {
--background: 0 0% 100%;
--foreground: 224 71.4% 4.1%;
/* Customize colors here */
}
```
### Modify Suggestion Buttons
Edit `components/ui/suggestion-buttons.tsx`:
```typescript
const suggestions = [
"Your custom suggestion 1",
"Your custom suggestion 2",
// Add more suggestions
]
```
## 📖 Advanced Usage
### Multiple Agents
Create different agents for different purposes:
```bash
# Create a coding assistant
LETTA_AGENT_TYPE=coding npm run setup-agent
# Create a research assistant
LETTA_AGENT_TYPE=research npm run setup-agent
```
### Custom Memory Blocks
Add domain-specific memory:
```typescript
const customMemory = {
name: "project_knowledge",
value: "Information about your specific project...",
description: "Domain-specific context and knowledge"
}
```
### Integration with External APIs
Extend the chat API to integrate with other services:
```typescript
// app/api/chat/route.ts
export async function POST(req: Request) {
// Add custom preprocessing
// Integrate with external APIs
// Custom response formatting
}
```
## 🐛 Troubleshooting
### Common Issues
| Problem | Solution |
|---------|----------|
| **Agent not responding** | Check API key in `.env.local` |
| **Styling issues** | Clear `.next` cache and restart |
| **Build errors** | Update dependencies: `npm update` |
| **Memory errors** | Restart development server |
### Debug Mode
Enable detailed logging:
```bash
DEBUG=letta* npm run dev
```
### Reset Agent
If your agent becomes unresponsive:
```bash
# Create a fresh agent
rm .env.local
cp env.example .env.local
# Add your API key and run setup again
npm run setup-agent
```
## 📊 Performance
- **First Load**: ~2s (with caching)
- **Message Response**: ~200ms average
- **Build Size**: ~500KB (gzipped)
- **Memory Usage**: ~50MB typical
## 🤝 Contributing
We welcome contributions! Please see our [Contributing Guide](CONTRIBUTING.md) for details.
### Development Workflow
1. Fork the repository
2. Create a feature branch: `git checkout -b feature/amazing-feature`
3. Make your changes with tests
4. Ensure code quality: `npm run lint`
5. Submit a pull request
### Code Style
- **TypeScript**: Strict mode enabled
- **ESLint**: Next.js recommended rules
- **Prettier**: Consistent code formatting
- **Conventional Commits**: Clear commit messages
## 📄 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## 🙏 Acknowledgments
- **[Letta Team](https://github.com/letta-ai/letta)** - For the amazing stateful agent platform
- **[shadcn](https://twitter.com/shadcn)** - For the beautiful UI component library
- **[Vercel](https://vercel.com/)** - For the excellent AI SDK and deployment platform
## 📞 Support
- **Documentation**: [Letta Docs](https://docs.letta.com/)
- **Community**: [Letta Discord](https://discord.gg/letta)
- **Issues**: [GitHub Issues](https://github.com/jyotirmi/letta-chatbot-iqagent/issues)
---
<div align="center">
**Built with ❤️ using [Letta](https://docs.letta.com/) and [Next.js](https://nextjs.org/)**
⭐ Star this repo if you found it helpful!
</div>
MCP 市场
