🤖 Claude Assistant — Full Stack AI Chat Application

A production-ready ChatGPT-like application built with Streamlit, FastAPI, and Claude API (Anthropic).

---

✨ Features

• 💬 Multi-turn conversations with thread-based memory
• 📁 File upload support (TXT, CSV, PDF)
• 🛠️ Function calling with built-in tools:
  • Weather information
  • Mathematical operations
  • Date/time queries
• 🎨 Clean, modern UI (ChatGPT-style)
• 🔒 Secure API communication
• ☁️ Deploy-ready for Streamlit Cloud

---

🏗️ Architecture

claude_assistant/
│
├── backend/              # FastAPI server
│   ├── main.py          # API endpoints
│   ├── tools.py         # Tool definitions
│   └── requirements.txt
│
├── frontend/            # Streamlit UI
│   ├── app.py          # Chat interface
│   └── requirements.txt
│
├── .env.example        # Environment template
├── Procfile           # Deployment config
└── README.md

---

🚀 Quick Start

Prerequisites

• Python 3.9+
• Anthropic API key (Get one here)

1. Clone Repository

git clone <your-repo-url>
cd claude_assistant

2. Set Up Environment

# Copy environment template
cp .env.example .env

# Add your API key to .env
echo "ANTHROPIC_API_KEY=your_actual_key_here" > .env

3. Install Dependencies

Backend:

cd backend
pip install -r requirements.txt

Frontend:

cd ../frontend
pip install -r requirements.txt

4. Run the Application

Terminal 1 — Start Backend:

cd backend
uvicorn main:app --reload --port 8000

Terminal 2 — Start Frontend:

cd frontend
streamlit run app.py

Access the app:

• Frontend: http://localhost:8501
• Backend API: http://localhost:8000
• API docs: http://localhost:8000/docs

---

☁️ Deploy to Streamlit Cloud

Option 1: Frontend Only (Simplified)

For Streamlit Cloud deployment, you can modify the frontend to call the Claude API directly:

1. Push your code to GitHub
2. Go to share.streamlit.io
3. Connect your repository
4. Add ANTHROPIC_API_KEY to Secrets (Settings → Secrets)
5. Deploy!

Note: For production with backend, deploy FastAPI separately (e.g., Railway, Render, Fly.io).

Option 2: Full Stack Deploy

1. Backend: Deploy to Railway/Render

   # Railway example
   railway init
   railway up

2. Frontend: Deploy to Streamlit Cloud

   • Set BACKEND_URL in Streamlit Secrets to your Railway/Render URL
   • Set ANTHROPIC_API_KEY in Secrets

---

🛠️ Available Tools

The assistant can execute these tools automatically:

Tool	Example Query
Weather	"What's the weather in Tokyo?"
Addition	"Add 42 and 17"
Multiplication	"Multiply 8 by 12"
Time	"What time is it?"

---

📝 API Endpoints

POST /chat

Main chat endpoint.

Request:

{
  "message": "Hello!",
  "thread_id": "optional-thread-id",
  "file_content": "optional file text",
  "file_name": "document.txt"
}

Response:

{
  "reply": "Hi! How can I help?",
  "thread_id": "uuid-here",
  "tool_used": "get_weather",
  "tool_result": "☀️ Sunny, 72°F"
}

GET /

Health check.

GET /tools

List available tools.

DELETE /thread/{thread_id}

Delete a conversation thread.

---

🔐 Environment Variables

Create a .env file:

# Required
ANTHROPIC_API_KEY=sk-ant-...

# Optional (for production)
BACKEND_URL=http://localhost:8000

For Streamlit Cloud, add these in Settings → Secrets:

ANTHROPIC_API_KEY = "sk-ant-..."
BACKEND_URL = "https://your-backend.railway.app"

---

📦 Project Dependencies

Backend:

• fastapi — Web framework
• anthropic — Claude API client
• uvicorn — ASGI server
• PyPDF2 — PDF processing

Frontend:

• streamlit — UI framework
• requests — HTTP client

---

🎯 Usage Examples

Basic Chat

User: Hello! How are you?
Claude: I'm doing well, thank you! How can I assist you today?

With Tools

User: What's the weather in London?
🛠️ Tool used: get_weather
Claude: The weather in London is currently 🌧️ Rainy, 55°F (13°C), with moderate wind.

With File Upload

User: [uploads data.csv] Summarize this data
Claude: I've analyzed your CSV file. It contains sales data with 3 columns...

---

🐛 Troubleshooting

Backend won't start:

# Check if port 8000 is in use
lsof -i :8000
# Use different port
uvicorn main:app --port 8001

"Cannot connect to backend" error:

• Ensure backend is running on port 8000
• Check BACKEND_URL in frontend
• Verify CORS settings in main.py

API key errors:

• Verify key is correctly set in .env
• Check key format starts with sk-ant-
• Ensure .env is in the correct directory

---

📄 License

MIT License — feel free to use in your own projects!

---

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

---

📧 Support

• Issues: Open a GitHub issue
• Anthropic API Docs: https://docs.anthropic.com
• Streamlit Docs: https://docs.streamlit.io

---

🎉 Credits

Built with ❤️ using:

• Anthropic Claude API
• FastAPI
• Streamlit

---

Ready to deploy? Just push to GitHub and connect to Streamlit Cloud! 🚀