🤖 AI Crypto Trading Bot

An intelligent cryptocurrency trading bot that uses AI analysis to make data-driven trading decisions on Coinbase Advanced Trade.

🎯 Core Philosophy: AI-First Trading

This bot prioritizes AI-driven market analysis over rigid portfolio rebalancing, allowing for:

Market-responsive trading based on technical indicators and trends
Dynamic position sizing that adapts to confidence levels and market conditions
Smart safety limits that protect your portfolio without blocking profitable trades
Momentum-based strategies that can ride winners and cut losers

✨ Key Features

🧠 AI-Powered Decision Making

Advanced LLM Analysis: Uses Google's Gemini 2.5 Pro for market analysis
Technical Indicators: RSI, MACD, Bollinger Bands, Moving Averages
Market Sentiment: Analyzes price trends, volume, and volatility
Confidence Scoring: 0-100% confidence levels for each trading decision

📊 Smart Trading Strategy

AI Priority: Market analysis drives all trading decisions
Dynamic Position Sizing: Trade sizes adapt to confidence (10-25% of balance)
Intelligent Rebalancing: Automatic portfolio rebalancing every 3 hours based on market conditions
Safety Limits: Minimum €50 EUR balance, 3% minimum crypto allocation

📱 Push Notifications

Real-time Trade Alerts: Instant notifications for BUY/SELL executions
Error Notifications: Critical error alerts with context
Portfolio Summaries: Daily portfolio performance updates
Bot Status Updates: Startup, shutdown, and status notifications

🛡️ Risk Management

Dual-Layer System: Static risk configuration + dynamic market assessment
Confidence Thresholds: 70% minimum confidence for BUY/SELL decisions
Position Sizing: Risk-adjusted trade sizes (HIGH=50% reduction, MEDIUM=25%, LOW=0%)
Market Volatility: Real-time assessment (high/medium/low) affects trading decisions
Balance Protection: Maintains minimum balances for continued trading
Diversification: Prevents over-concentration in any single asset

📚 Complete Risk Management Guide →

📱 Professional Dashboard

Real-time Portfolio Tracking: Live EUR values and allocations
Comprehensive AI Analysis: Detailed reasoning for each trade decision with 4 analysis sections
Performance Metrics: P&L tracking, win rates, portfolio growth
Market Overview: Current prices, trends, and technical indicators

🧪 Test Coverage & Quality Assurance

📊 Test Implementation Progress

Phase 1 (Critical Components):     ████████████████████ 100% (5/5)
Phase 2 (Core Functionality):     ████████░░░░░░░░░░░░  20% (1/5)
Phase 3 (Integration & E2E):      ░░░░░░░░░░░░░░░░░░░░  0% (0/5)
Phase 4 (Specialized Testing):    ░░░░░░░░░░░░░░░░░░░░  0% (0/7)

Overall Test Suite Progress:       ████████████░░░░░░░░ 31.8% (7/22)

✅ Completed Tests

✅ Configuration Management (test_config.py) - 100% coverage - 37 tests
- Environment variable parsing and validation
- Trading pairs and allocation calculations
- Risk management settings
- Backward compatibility
✅ Coinbase Client (test_coinbase_client.py) - 53% coverage - 38 tests
- Client initialization and credential validation
- Rate limiting and error handling
- Account operations and balance retrieval
- Market data operations and price fetching
- Trading operations and order placement
- Precision handling for different trading pairs
- Notification integration and error recovery
✅ Trading Strategy (test_trading_strategy.py) - Core functionality - 14 tests
- Strategy initialization and configuration
- Trading decision logic with technical indicators
- Risk management and confidence thresholds
- Position sizing calculations
- Portfolio management operations
- Error handling and edge cases
✅ Risk Management (test_risk_management.py) - Comprehensive coverage - 32 tests
- Risk level configuration and multipliers
- Confidence threshold enforcement
- Balance and safety checks
- Smart trading limits and portfolio protection
- Position sizing with risk management
- Confidence adjustments based on technical indicators
- Integrated risk scenarios and error handling
✅ Portfolio Management (test_portfolio.py) - Complete coverage - 41 tests
- Portfolio initialization and loading from multiple sources
- Portfolio validation and data structure integrity
- Portfolio saving and persistence across instances
- Portfolio value calculations and price updates
- Asset allocation calculations and performance metrics
- Trade execution and portfolio updates
- Exchange data synchronization and integration
- Portfolio rebalancing calculations and actions
- Error handling and edge case management
- Large number precision and concurrent access safety
- Rate limiting and error handling
- Account operations and balance retrieval
- Market data operations and price fetching
- Trading operations and order placement
- Precision handling for different trading pairs
- Notification integration and error recovery
✅ LLM Analyzer (test_llm_analyzer.py) - 100% coverage - 26 tests
- Initialization and Google Cloud authentication
- Vertex AI provider integration and API calls
- Market data analysis and decision generation
- Technical indicator processing and prompt creation
- Response parsing and JSON handling
- Trading decision making (BUY/SELL/HOLD signals)
- Error handling and edge cases
- Configuration validation and parameter handling

🎉 Phase 1 Complete!

All critical components now have comprehensive test coverage.

🎉 LLM Analyzer Complete!

First Phase 2 component with 100% test coverage and 26 comprehensive tests.

📋 Next Priority Tests (Phase 2)

📋 Data Collector (test_data_collector.py)
📋 Dashboard Updater (test_dashboard_updater.py)
📋 Notification Service (test_notification_service.py)
📋 Trade Logger (test_trade_logger.py)

🎯 Quality Metrics

Code Coverage Target: 90%+ for unit tests
Current Coverage: 100% (config), 53% (coinbase_client), 100% (llm_analyzer), Comprehensive (core modules)
Test Execution Time: <1 second per module
Test Categories: Unit, Integration, E2E, Performance, Security

📚 Complete Test Documentation →

🚀 Quick Start

Prerequisites

Python 3.8+
Coinbase Advanced Trade account
Google Cloud account (for AI analysis)

Installation

Clone the repository

git clone <repository-url>
cd AI-crypto-bot

Create virtual environment

python3 -m venv venv
source venv/bin/activate  # Linux/Mac
# or
venv\Scripts\activate     # Windows

Install dependencies
```
pip install -r requirements.txt
```

Configure environment

cp .env.example .env
# Edit .env with your API keys and settings

Deploy dashboard (optional)
```
python3 deploy_dashboard.py
```
Run the bot
```
python3 main.py
```

☁️ Cloud Deployment

For production deployment on cloud platforms, use the automated setup scripts:

AWS EC2: See aws_setup/README.md for complete AWS deployment guide
Google Cloud: See gcp_setup/README.md for complete GCP deployment guide

Both include automated scripts for dependencies, web server setup, and service configuration.

⚙️ Configuration

Core Settings (`.env`)

# Trading Configuration
BASE_CURRENCY=EUR                    # Base currency for trading
TRADING_PAIRS=BTC-EUR,ETH-EUR,SOL-EUR  # Trading pairs
RISK_LEVEL=HIGH                      # LOW, MEDIUM, HIGH
SIMULATION_MODE=False                # Set to True for paper trading

# AI Decision Thresholds
CONFIDENCE_THRESHOLD_BUY=70          # Minimum confidence for BUY orders
CONFIDENCE_THRESHOLD_SELL=70         # Minimum confidence for SELL orders

# Trade Size Limits
MIN_TRADE_AMOUNT=30.0               # Minimum trade size (EUR)
MAX_POSITION_SIZE=1000.0            # Maximum single trade size (EUR)

# Portfolio Strategy
TARGET_CRYPTO_ALLOCATION=90         # Target crypto allocation %
TARGET_BASE_ALLOCATION=10           # Target EUR allocation %

API Configuration

# Coinbase Advanced Trade
COINBASE_API_KEY=your_api_key
COINBASE_API_SECRET=your_private_key

# Google Cloud (for AI analysis)
GOOGLE_CLOUD_PROJECT=your_project_id
GOOGLE_APPLICATION_CREDENTIALS=path/to/service-account.json
LLM_MODEL=gemini-2.5-pro-preview-05-06

# Push Notifications (Pushover)
NOTIFICATIONS_ENABLED=true
PUSHOVER_TOKEN=your_pushover_app_token
PUSHOVER_USER=your_pushover_user_key

Push Notification Setup

Download Pushover App
- iOS: App Store
- Android: Google Play
Create Pushover Account
- Sign up at pushover.net
- Note your User Key from the dashboard
Create Application
- Go to Create Application
- Name: "AI Crypto Bot" (or your preference)
- Copy the API Token/Key

Configure Environment

NOTIFICATIONS_ENABLED=true
PUSHOVER_TOKEN=your_app_api_token_here
PUSHOVER_USER=your_user_key_here

Test Notifications
```
python3 test_notifications.py
```

🎛️ Trading Strategy Details

AI-First Approach

Unlike traditional rebalancing bots that force rigid allocation percentages, this bot:

Analyzes Market Conditions using AI and technical indicators
Makes Trading Decisions based on market opportunities
Applies Smart Safety Limits to protect your portfolio
Sizes Positions Dynamically based on confidence and allocation

Smart Safety Limits

Minimum EUR Balance: €50 (ensures ability to make BUY orders)
Maximum EUR Allocation: 35% (prevents holding too much cash)
Minimum Crypto Allocation: 3% per asset (maintains diversification)
Trade Size Scaling: Higher confidence = larger positions (up to 1.5x)

Decision Process

Market Data → AI Analysis → Confidence Score → Safety Check → Position Sizing → Execute Trade

📊 Dashboard Features

Portfolio Overview

Total Value: Real-time EUR portfolio value
Asset Allocation: Current vs target percentages
AI-First Strategy Status: Smart safety limits and trading capability
Performance Metrics: P&L, returns, win rates

Enhanced AI Analysis

Comprehensive Trade Details: 4-section analysis for each trade
- AI Reasoning: Detailed bullet-point market analysis
- Technical Analysis: RSI, MACD, Bollinger Bands with color coding
- Bot Decision Process: Confidence checks and risk management
- Market Context: Price trends, conditions, and analysis quality
Recent Decisions: Latest BUY/SELL/HOLD recommendations
Confidence Levels: AI confidence scores (0-100%)

Live Logs Dashboard

Real-time Log Monitoring: Last 30 lines of bot activity
Auto-refresh: Updates every 5 seconds automatically
Color-coded Log Levels: ERROR (red), WARNING (yellow), INFO (blue), DEBUG (gray)
Live Status Indicators: Bot operational status and last update time
Interactive Controls: Manual refresh, auto-scroll toggle, clear logs

Market Data

Live Prices: Real-time cryptocurrency prices
Price Changes: 1h, 4h, 24h, 5d percentage changes
Trading Status: Current bot status and next decision time

🏗️ Project Structure

AI-crypto-bot/
├── main.py                 # Main bot entry point
├── config.py              # Configuration management
├── coinbase_client.py     # Coinbase API client
├── llm_analyzer.py        # AI analysis engine
├── trading_strategy.py    # Trading strategy implementation
├── data_collector.py      # Market data collection
├── utils/                 # Utility modules
│   ├── dashboard_updater.py
│   ├── webserver_sync.py
│   ├── logger.py
│   ├── portfolio.py
│   └── ...
├── dashboard/             # Web dashboard
│   ├── static/
│   └── images/
├── aws_setup/             # AWS deployment scripts
│   ├── setup_ec2.sh
│   ├── iam_policy.json
│   ├── cloudwatch_setup.sh
│   └── create_secrets.sh
├── gcp_setup/             # Google Cloud deployment scripts
│   ├── setup_gce.sh
│   └── README.md
├── data/                  # Runtime data
├── logs/                  # Application logs
├── .env                   # Environment variables
└── requirements.txt       # Dependencies

🕒 Persistent Uptime Tracking

The bot features intelligent uptime management that distinguishes between restarts and stops:

Uptime Behavior

Restarts: Configuration changes, updates, or service restarts preserve uptime
Explicit Stops: Manual stops or systemctl stop reset uptime
Crash Recovery: Automatic restarts after crashes preserve uptime

Bot Management

Use the included bot manager for proper uptime handling:

# Start the bot
python3 bot_manager.py start

# Restart (preserves uptime) - for config changes, updates
python3 bot_manager.py restart

# Stop (resets uptime) - for maintenance, shutdown
python3 bot_manager.py stop

# Check status and total uptime
python3 bot_manager.py status

Dashboard Display

Total Uptime: Cumulative uptime across all sessions
Restart Count: Number of restarts since original start
Professional Metrics: True operational uptime tracking

🔧 Advanced Configuration

Risk Management

# Position Sizing Multipliers
RISK_HIGH_POSITION_MULTIPLIER=0.5    # 50% position reduction for high risk
RISK_MEDIUM_POSITION_MULTIPLIER=0.75 # 25% position reduction for medium risk
RISK_LOW_POSITION_MULTIPLIER=1.0     # No reduction for low risk

# Technical Analysis Weights
RSI_SIGNAL_WEIGHT=0.3               # RSI indicator weight
MACD_SIGNAL_WEIGHT=0.4              # MACD indicator weight
BOLLINGER_SIGNAL_WEIGHT=0.3         # Bollinger Bands weight

Dashboard Customization

# Web Interface
WEBSERVER_SYNC_ENABLED=true         # Enable web dashboard
WEBSERVER_SYNC_PATH=/var/www/html/crypto-bot  # Dashboard path
DASHBOARD_TRADE_HISTORY_LIMIT=10    # Number of recent trades to show

📈 Performance Optimization

Confidence-Based Trading

High Confidence (80%+): Larger position sizes, immediate execution
Medium Confidence (70-79%): Standard position sizes
Low Confidence (<70%): No trade execution (HOLD)

Dynamic Position Sizing

Over-allocated assets: Smaller SELL positions to maintain balance
Under-allocated assets: Larger BUY positions when EUR is available
High EUR allocation: Larger BUY positions to deploy cash
Low EUR allocation: Smaller BUY positions to preserve liquidity

🛠️ Maintenance & Monitoring

Log Files

Main Log: logs/crypto_bot.log - General bot activity
Trade Log: data/trades/trade_history.json - All executed trades
Decision Log: data/cache/latest_decisions.json - Recent AI decisions

Data Files

Portfolio: data/portfolio/portfolio.json - Current holdings
Market Data: data/*_EUR_*.json - Historical price and analysis data
Configuration: data/config/ - Runtime configuration snapshots

Health Checks

Dashboard Status: Green badge indicates healthy operation
EUR Balance: Warning if below €50 minimum
API Connectivity: Automatic retry on connection issues
AI Analysis: Fallback to HOLD if analysis fails

🚨 Safety Features

Simulation Mode

Test strategies without real money:

SIMULATION_MODE=True

Emergency Stops

Insufficient Balance: Stops BUY orders if EUR < €50
API Failures: Graceful degradation with error logging
Invalid Decisions: Defaults to HOLD on analysis errors

Portfolio Protection

Minimum Holdings: Prevents selling below 3% allocation
Maximum Cash: Limits EUR allocation to 35%
Trade Size Limits: €30 minimum, €1000 maximum per trade

🔄 Deployment

Local Deployment

Systemd Service

# Copy service file
sudo cp crypto-bot.service /etc/systemd/system/
sudo systemctl daemon-reload
sudo systemctl enable crypto-bot
sudo systemctl start crypto-bot

Dashboard Deployment

# Deploy dashboard to web server
python3 deploy_dashboard.py

☁️ Cloud Deployment

🚀 AWS EC2 Deployment

1. Launch EC2 Instance

Choose Amazon Linux 2023 AMI
Select t2.medium or better for production
Configure security groups:
- SSH access (port 22)
- HTTP access (port 80) for dashboard

2. Connect and Setup

# Connect to instance
ssh -i your-key.pem ec2-user@your-instance-ip

# Clone repository
git clone https://github.com/schmidb/AI-crypto-bot.git
cd AI-crypto-bot

# Run setup script
bash aws_setup/setup_ec2.sh

3. Configure and Start

# Edit configuration
nano .env

# Start service
sudo systemctl start crypto-bot
sudo systemctl enable crypto-bot

# Check status
sudo systemctl status crypto-bot
sudo journalctl -u crypto-bot -f

4. Access Dashboard

http://your-ec2-ip/crypto-bot/

🌐 Google Cloud Deployment

1. Create VM Instance

Navigate to Compute Engine > VM instances
Choose e2-medium or better
Select Debian/Ubuntu boot disk
Allow HTTP/HTTPS traffic
Add firewall rule for port 80

2. Connect and Setup

# Connect via SSH
gcloud compute ssh your-instance-name --zone=your-zone

# Clone repository
git clone https://github.com/schmidb/AI-crypto-bot.git
cd AI-crypto-bot

# Run setup script
bash gcp_setup/setup_gce.sh

3. Configure and Start

# Edit configuration
nano .env

# Check service status
sudo supervisorctl status crypto-bot

# View logs
tail -f /var/log/crypto-bot/crypto-bot.log

# Restart if needed
sudo supervisorctl restart crypto-bot

4. Access Dashboard

http://your-vm-ip/crypto-bot/

🔐 Security Configuration

AWS IAM Policy (for CloudWatch integration):

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "cloudwatch:PutMetricData",
        "cloudwatch:GetMetricStatistics",
        "logs:CreateLogGroup",
        "logs:CreateLogStream",
        "logs:PutLogEvents"
      ],
      "Resource": "*"
    }
  ]
}

GCP Firewall Rules:

# Allow HTTP traffic for dashboard
gcloud compute firewall-rules create allow-http-crypto-bot \
    --allow tcp:80 \
    --source-ranges 0.0.0.0/0 \
    --description "Allow HTTP access to crypto bot dashboard"

📞 Support & Troubleshooting

Common Issues

Bot not trading:

Check EUR balance (minimum €50 required)
Verify AI confidence meets 70% threshold
Confirm API keys are valid

Dashboard not updating:

Check webserver sync settings
Verify file permissions
Restart bot if needed

Low performance:

Review confidence thresholds
Check market conditions
Analyze recent AI decisions in dashboard

Getting Help

Check log files for error messages
Review dashboard for status indicators
Verify configuration settings
Test in simulation mode first

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

⚠️ Disclaimer

This software is for educational and research purposes. Cryptocurrency trading involves substantial risk of loss. Past performance does not guarantee future results. Use at your own risk and never invest more than you can afford to lose.

Built with ❤️ for intelligent cryptocurrency trading

Name		Name	Last commit message	Last commit date
Latest commit History 177 Commits
.github/workflows		.github/workflows
aws_setup		aws_setup
dashboard		dashboard
examples		examples
gcp_setup		gcp_setup
strategies		strategies
tests		tests
utils		utils
.coverage		.coverage
.env.example		.env.example
.gitignore		.gitignore
DAY_TRADING_OPTIMIZATION.md		DAY_TRADING_OPTIMIZATION.md
DEVELOPMENT_PLAN.md		DEVELOPMENT_PLAN.md
MIGRATION_TO_MULTI_STRATEGY.md		MIGRATION_TO_MULTI_STRATEGY.md
README.md		README.md
RISK_MANAGEMENT.md		RISK_MANAGEMENT.md
STRUCTURE.md		STRUCTURE.md
UPTIME_MANAGEMENT.md		UPTIME_MANAGEMENT.md
add_multi_strategy.py		add_multi_strategy.py
bot_manager.py		bot_manager.py
coinbase_client.py		coinbase_client.py
config.py		config.py
crypto-bot.service		crypto-bot.service
data_collector.py		data_collector.py
deploy_dashboard.py		deploy_dashboard.py
google.json.template		google.json.template
llm_analyzer.py		llm_analyzer.py
llm_prompt_enhancement.py		llm_prompt_enhancement.py
main.py		main.py
pytest.ini		pytest.ini
requirements-dev.txt		requirements-dev.txt
requirements.txt		requirements.txt
set_performance_goals.py		set_performance_goals.py
trading_strategy.py		trading_strategy.py

schmidb/AI-crypto-bot

Folders and files

Latest commit

History

Repository files navigation

🤖 AI Crypto Trading Bot

🎯 Core Philosophy: AI-First Trading

✨ Key Features

🧠 AI-Powered Decision Making

📊 Smart Trading Strategy

📱 Push Notifications

🛡️ Risk Management

📱 Professional Dashboard

🧪 Test Coverage & Quality Assurance

📊 Test Implementation Progress

✅ Completed Tests

🎉 Phase 1 Complete!

🎉 LLM Analyzer Complete!

📋 Next Priority Tests (Phase 2)

🎯 Quality Metrics

🚀 Quick Start

Prerequisites

Installation

☁️ Cloud Deployment

⚙️ Configuration

Core Settings (.env)

API Configuration

Push Notification Setup

🎛️ Trading Strategy Details

AI-First Approach

Smart Safety Limits

Decision Process

📊 Dashboard Features

Portfolio Overview

Enhanced AI Analysis

Live Logs Dashboard

Market Data

🏗️ Project Structure

🕒 Persistent Uptime Tracking

Uptime Behavior

Bot Management

Dashboard Display

🔧 Advanced Configuration

Risk Management

Dashboard Customization

📈 Performance Optimization

Confidence-Based Trading

Dynamic Position Sizing

🛠️ Maintenance & Monitoring

Log Files

Data Files

Health Checks

🚨 Safety Features

Simulation Mode

Emergency Stops

Portfolio Protection

🔄 Deployment

Local Deployment

Systemd Service

Dashboard Deployment

☁️ Cloud Deployment

🚀 AWS EC2 Deployment

🌐 Google Cloud Deployment

🔐 Security Configuration

📞 Support & Troubleshooting

Common Issues

Getting Help

📄 License

⚠️ Disclaimer

About

Topics

Resources

Uh oh!

Stars

Watchers

Forks

Releases 1

Languages

Core Settings (`.env`)