A comprehensive Python Flask web application that combines real-time stock data, AI-powered trading recommendations, automated portfolio management, and performance tracking. Features include intelligent stock analysis, automated trading execution, and advanced portfolio monitoring.
Note: this is not an offer to buy or sell stock. This is was built as an experiment to see if ChatGPT could trade or not.
I don't endorse anyone using this to manage thier portfolo. It is a good example of testing out some APIs including Open AI API
Please invest with a professional! This does not represent my current, past or future employment.
All the code was written by AI (incluing the prompt to AI ) as this was also used as a demo to how to vibe code a system.
This was in response to https://github.com/LuckyOne7777/ChatGPT-Micro-Cap-Experiment trading success.
- Multi-API stock quote system with automatic failover
- 1-hour cached quotes system for performance optimization
- Support for 5+ free stock APIs with rotation
- Real-time price updates and market data
- OpenAI GPT-4 integration for intelligent stock analysis
- Automated buy/sell recommendations with reasoning
- Risk assessment and stop-loss price suggestions
- Market sentiment analysis and trend prediction
- Complete portfolio tracking with CSV-based storage
- Buy/sell transaction execution with P&L calculations
- Holdings management with weighted average cost basis
- Cash balance tracking and transaction history
- Interactive Chart.js performance visualization with S&P 500 comparison
- Daily performance tracking with persistent CSV storage
- Portfolio vs. market benchmark analysis (SPY ETF)
- Percentage-based gain/loss charts with baseline normalization
- Real-time performance metrics and stop-loss monitoring
- 9:30 AM Weekdays: AI predictor with automatic trade execution
- 10 AM, 12 PM, 2 PM, 3:30 PM: Stop-loss monitoring and protection
- Comprehensive logging and error handling
- Weekday-only operation with market hours awareness
- Bootstrap-powered responsive design
- Tabbed interface for different portfolio functions
- Real-time data updates and interactive charts
- Mobile-friendly and professional UI
- yfinance - No API key required, reliable data source
- Alpha Vantage - 25 requests/day free tier
- pyfinviz - Financial data scraping
- Finnhub - 60 calls/minute free tier
- Free APIs - Various backup data sources
-
Clone the repository and navigate to the project directory
-
Install Python dependencies:
pip install -r requirements.txt- Set up environment variables (.env): Copy the sample file and edit your values:
cp .env_sample .envThen open .env and set at minimum:
OPENAI_API_KEY=your_openai_api_key_here
ALPHA_VANTAGE_API_KEY=your_alpha_vantage_key_here
FINHUB_API_KEY=your_finnhub_key_here
# Email (Gmail)
USERID=your_gmail_address@gmail.com
APP_PASSWORD=your_16_char_gmail_app_password
Notes:
- Create a Gmail App Password in Google Account → Security → 2‑Step Verification → App passwords.
- The scheduler auto-detects Gmail and uses
smtp.gmail.com:465(SSL) andimap.gmail.comby default. - You may optionally override SMTP/IMAP via env:
SMTP_HOST,SMTP_PORT,SMTP_USE_SSL,IMAP_HOST,IMAP_CHECK.
- Run the web application:
python app.py- Open your browser and navigate to
http://localhost:5000
python scheduler.pypython test_scheduler.pyThe scheduler will automatically:
- Execute AI-driven trades at 9:30 AM on weekdays
- Monitor stop-losses at 10 AM, 12 PM, 2 PM, and 3:30 PM
- Update cached quotes and log all activities
GET /api/quote/<symbol>- Get current stock quoteGET /api/historical/<symbol>?period=1mo- Get historical dataGET /api/multiple?symbols=AAPL,GOOGL,TSLA- Get multiple quotes
POST /api/portfolio/buy- Execute buy orderPOST /api/portfolio/sell- Execute sell orderGET /api/portfolio/summary- Get portfolio overviewGET /api/portfolio/transactions- Get transaction historyGET /api/portfolio/performance- Get performance chart dataPOST /api/portfolio/refresh-quotes- Override quote cache
POST /api/ai/recommendations- Get AI trading recommendationsPOST /api/ai/execute-trades- Execute selected AI recommendations
The system uses CSV files for data persistence:
data/transactions.csv- All buy/sell transactions with P&L and stop-loss pricesdata/holdings.csv- Current portfolio holdings with market valuesdata/cash.csv- Cash balance trackingdata/cached_quotes.csv- 1-hour cached stock quotes including SPYdata/daily_performance.csv- Daily portfolio and S&P 500 performance tracking
- Navigate to the Portfolio tab
- Use Buy/Sell forms to execute trades
- View performance charts and transaction history
- Monitor real-time portfolio value
- Go to the AI Predictor tab
- Click "Get AI Recommendations"
- Review AI analysis and suggestions
- Select trades to execute or let scheduler handle automatically
- Start the scheduler:
python scheduler.py - The system will automatically:
- Run AI analysis at market open (9:30 AM)
- Execute recommended trades
- Monitor stop-losses throughout the day
- Log all activities to
scheduler.log
Environment is loaded from .env (see Installation step). Core variables:
# APIs
OPENAI_API_KEY=your_openai_api_key_here
ALPHA_VANTAGE_API_KEY=your_alpha_vantage_key_here
FINHUB_API_KEY=your_finnhub_key_here
# Email (Gmail)
USERID=your_gmail_address@gmail.com
APP_PASSWORD=your_16_char_gmail_app_password
# Optional SMTP/IMAP overrides
# SMTP_HOST=smtp.gmail.com
# SMTP_PORT=465
# SMTP_USE_SSL=true
# IMAP_HOST=imap.gmail.com
# IMAP_CHECK=true
# SMTP_DEBUG=false
- The app prefers the
certifiCA bundle for SSL.requirements.txtincludescertifi. - If you need to bypass IMAP pre-check, set
IMAP_CHECK=false.
The system starts with $10,000 in cash. Modify in portfolio_service.py if needed.
Modify scheduler.py to adjust trading times and frequency.
The system automatically creates and maintains a daily_performance.csv file with the following structure:
date,portfolio_value,portfolio_gain_loss,portfolio_gain_loss_pct,spy_price,spy_gain_loss,spy_gain_loss_pct
2025-09-17,9989.2,-10.80,-0.108,659.18,-0.12,-0.018Key Features:
- Automatic daily updates when accessing performance charts
- Baseline tracking from $10,000 initial portfolio value
- SPY benchmark comparison using cached quote system
- Persistent storage for historical performance analysis
- Accurate percentage calculations for portfolio vs. market performance
- Automated daily performance recording in
daily_performance.csv - Portfolio vs. S&P 500 comparison using SPY ETF as benchmark
- Baseline normalization from $10,000 starting portfolio
- Percentage-based performance charts for accurate comparison
- Persistent performance history with CSV storage
- Real-time portfolio tracking with gain/loss calculations
- Interactive dual-line charts (Portfolio vs. Market)
- Stop-loss monitoring with risk percentage display
- Color-coded metrics (green for gains, red for losses)
- Transaction history with detailed P&L and stop-loss tracking
- SPY ETF integration for S&P 500 market comparison
- Cached quote system for efficient data retrieval
- Day-over-day performance calculation and display
- Risk assessment showing downside protection percentages
The application automatically rotates through multiple stock APIs if one fails, ensuring reliable data access.
1-hour cached quotes system reduces API calls while maintaining data freshness.
OpenAI GPT-4 analyzes market conditions, company fundamentals, and provides reasoned trading recommendations.
- Automatic stop-loss price suggestions
- Cash balance validation before trades
- Portfolio diversification tracking
- Web Interface: Real-time portfolio monitoring
- Scheduler Logs: Detailed trading activity logs in
scheduler.log - Console Output: Live trading updates and status messages
- Error Handling: Comprehensive error logging and recovery
The modular architecture makes it easy to add:
- Additional stock APIs
- New AI models or trading strategies
- Custom technical indicators
- Advanced portfolio analytics
- Integration with brokers' APIs
- Mobile app connectivity