Cloud Data Analyst

AI-powered data analysis platform using Claude and Modal sandboxes.

Architecture

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│   Next.js       │────▶│   FastAPI       │────▶│   Modal         │
│   Frontend      │     │   Backend       │     │   Sandbox       │
└─────────────────┘     └─────────────────┘     └─────────────────┘
                              │                        │
                              ▼                        ▼
                        ┌─────────────────┐     ┌─────────────────┐
                        │   Claude API    │     │   Modal Volume  │
                        └─────────────────┘     │   (Shared)      │
                                                └─────────────────┘

Setup

Prerequisites

• Python 3.11+
• Node.js 18+
• Modal account (https://modal.com)
• Anthropic API key

Backend Setup

cd backend

# Create virtual environment
python -m venv venv
source venv/bin/activate  # or `venv\Scripts\activate` on Windows

# Install dependencies
pip install -r requirements.txt

# Set environment variables
export ANTHROPIC_API_KEY=your-api-key
export MODAL_TOKEN_ID=your-modal-token-id
export MODAL_TOKEN_SECRET=your-modal-token-secret

# Deploy Modal app (first time only)
modal deploy modal_sandbox.py

# Start the backend
uvicorn main:app --reload

Frontend Setup

cd frontend

# Install dependencies
npm install

# Start development server
npm run dev

Usage

1. Open http://localhost:3000
2. Upload a CSV file using the drag-drop zone
3. Ask questions about your data:
   • "What's in this file?"
   • "Show me summary statistics"
   • "Create a histogram of [column]"
   • "What correlations exist between columns?"

Features

• Secure Execution: Code runs in Modal sandboxes with no network access
• Data Analysis: pandas, numpy, matplotlib, seaborn pre-installed
• Streaming Responses: Real-time streaming from Claude
• Visualization: Matplotlib plots displayed inline

Security

• 60-second execution timeout
• Network disabled in sandboxes
• Path validation prevents directory traversal
• 10MB file upload limit
• CSV, JSON, TXT files only
• 24-hour session expiry

API Endpoints

• POST /api/sessions - Create new session
• POST /api/sessions/{id}/upload - Upload file
• POST /api/sessions/{id}/chat - Chat with Claude (SSE)
• GET /api/sessions/{id}/files - List files
• GET /api/sessions/{id}/files/{name} - Download file