GrindhouseBot 🤖

A powerful Discord bot for cryptocurrency trading signals and real-time market monitoring using the Bybit exchange. The bot provides automated technical analysis signals, market insights, and AI-powered chat assistance directly in your Discord channel.

Features 🌟

• Real-time trading signals using multiple technical analysis strategies
• AI-enhanced signal analysis
• Dynamic market analysis with adaptive thresholds
• Live monitoring of USDT perpetual contracts
• Market performance tracking with top gainers and losers
• Multilingual AI chat assistance
• Efficient in-memory price data management
• Docker support for easy deployment
• Extensible strategy system with factory pattern
• Adjustable time intervals for analysis
• Multi-strategy signal monitoring
• Combined confidence scoring system

Signalling Strategies 📈

RSI Strategy

• Oversold/Overbought detection
• Confidence based on RSI extremes
• RSI < 30: Buy signal (stronger as RSI decreases)
• RSI > 70: Sell signal (stronger as RSI increases)

MACD Strategy

• Trades divergence extremes with adaptive thresholds
• Buy signals on significant negative divergence (maximum fear)
• Sell signals on significant positive divergence (maximum greed)
• Dynamic threshold calculation using median absolute deviation
• Automatically adapts to different timeframes and market conditions

Bollinger Bands Strategy

• Band penetration signals
• Confidence based on penetration depth
• Normalized by band width
• Adapts to market volatility

Ichimoku Strategy

• Cloud position signals
• TK cross confirmation
• Weighted confidence calculation:
  • 40%: Distance from cloud
  • 30%: TK cross strength
  • 30%: Cloud thickness

Signal Confidence

Signals combine multiple factors:

1. Individual strategy confidence (0.0-1.0)
2. Strategy agreement ratio
3. Dynamic aggregation based on active strategies

Strategy-Specific Confidence Scoring

Each strategy uses optimized confidence calculations:

Bollinger Bands

• Base confidence from band penetration depth
• Adjusts for current volatility using band width
• Considers distance from middle band for trend context
• Weighted scoring: 70% band penetration + 30% trend context

RSI

• Base confidence from distance to extremes
• Includes momentum (rate of change) consideration
• Bonus confidence for extreme levels (RSI < 20 or > 80)
• Weighted scoring: 60% RSI level + 40% momentum, with up to 10% extreme bonus

MACD

• Base confidence from divergence magnitude vs dynamic threshold
• Considers histogram strength relative to recent movements
• Includes trend consistency factor
• Weighted scoring:
  • 50% divergence magnitude
  • 30% histogram strength
  • 20% trend consistency
  • +10% bonus for extreme divergences
• Uses median absolute deviation for adaptive thresholds
• Minimum sensitivity threshold to prevent noise

Ichimoku

• Measures cloud and TK cross relative to recent price range
• Considers cloud thickness for trend strength
• Weighted scoring: 45% price position + 35% TK cross strength + 20% cloud strength

Signal Aggregation

The signal handler implements an adaptive approach for combining signals:

1. Dynamic Confidence Calculation

   • Final confidence combines two key factors:
     • Average confidence of agreeing signals
     • Ratio of agreeing strategies vs active strategies
   • Example:
     • If listening to 3 strategies (RSI, MACD, Bollinger)
     • And 2 strategies agree with average confidence 0.8
     • Final confidence = 0.8 * (2/3) = 0.53

2. Advantages of This Approach

   • Adapts automatically as strategies are added/removed
   • No fixed weights, making it easy to add new strategies
   • Higher confidence when more active strategies agree
   • Naturally scales with the number of strategies in use

3. Signal filtering

   • Only aggregated signals higher than average confidence score are displayed
   • Displays only useful aggregated signals

Performance Optimizations 🚀

Technical Indicators

• Uses pandas_ta for efficient indicator calculations
  • Optimized with Cython for high performance
  • Efficient memory usage for rolling windows
  • Vectorized operations for speed

Signal Processing

• Numpy-based confidence calculations
• Efficient signal aggregation using vectorized operations
• Pre-calculated lookup tables for weights and templates
• Optimized string formatting for Discord messages

Memory Management

• Efficient DataFrame management
• Smart window sizes for each strategy
• Minimal data copying in signal processing
• Optimized message batching for Discord

Installation 🔧

Prerequisites

1. Create a Discord bot using the Discord Developer Portal

   • Enable "Message Content Intent" in the bot settings
   • Generate a bot token
   • Add the bot to your server with necessary permissions

2. Get an OpenRouter API key from OpenRouter

   • Sign up for an account
   • Generate an API key
   • Add the key to your .env file

Standard Installation

1. Clone the repository
2. Create a virtual environment:

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

3. Install dependencies:

   pip install -r requirements.txt
   pip install -U git+https://github.com/twopirllc/pandas-ta.git@development

4. Create a .env file with your credentials:

   DISCORD_BOT_TOKEN=your_discord_bot_token
   OPENROUTER_API_KEY=your_openrouter_api_key
   DEFAULT_INTERVAL=60
   

Docker Installation

1. Build and run using Docker Compose:

   docker compose up --build

Time Interval Configuration ⏰

The DEFAULT_INTERVAL setting in your .env file determines the timeframe for price analysis. Valid intervals are:

• Minutes: '1', '3', '5', '15', '30', '60', '120', '240', '360', '720'
• Daily: 'D' (1440 minutes)
• Weekly: 'W' (10080 minutes)
• Monthly: 'M' (43200 minutes)

Choose an interval that matches your trading strategy. For example:

• Short-term trading: Use shorter intervals (1-15 minutes)
• Swing trading: Use medium intervals (60-240 minutes)
• Long-term analysis: Use longer intervals (D/W/M)

Commands 💬

• !listen <strategies> - Start monitoring for trading signals

  • Available strategies: rsi, macd, bollinger, ichimoku
  • Use comma-separated values for multiple strategies: !listen rsi,macd,bollinger
  • Use all to enable all strategies: !listen all
  • Examples:

    !listen rsi                         # Single strategy
    !listen rsi,macd,bollinger         # Multiple strategies
    !listen all                        # All available strategies
    

• !unlisten <strategy> - Stop monitoring a specific strategy

  • Use without strategy to stop all monitoring
  • Examples:

    !unlisten rsi      # Stop specific strategy
    !unlisten          # Stop all strategies
    

• !interval <value> - Change the analysis timeframe

  • Valid intervals:
    • Minutes: 1, 3, 5, 15, 30, 60, 120, 240, 360, 720
    • Special: D (daily), W (weekly), M (monthly)
  • Examples:

    !interval 240  # Switch to 4-hour candles
    !interval 60   # Switch to 1-hour candles
    !interval D    # Switch to daily candles
    

• !chat <question> - Ask the bot about crypto trading or its features

  • Multilingual support
  • Get concise answers about:
    • Cryptocurrency markets and trading
    • Bot features and strategies
    • Technical analysis concepts
  • Examples:

    !chat How does the RSI strategy work?
    !chat What is a Bollinger Band squeeze?
    !chat Explain the Ichimoku cloud settings
    

• !top winners - Display top 5 performing coins in the last 24 hours

• !top losers - Display worst 5 performing coins in the last 24 hours

• !clear <count> - Clear specified number of messages from the channel

  • Example: !clear 100

Signal Output Format 📊

The bot provides detailed, actionable trading signals:

🎯 High-Confidence Trading Signals

📈 BUY Signals
**BTCUSDT** (0.85)
Price: 43250
RSI 28 | MACD Div -1.20% | BB 2.50%

📉 SELL Signals
**ETHUSDT** (0.92)
Price: 2250
RSI 75 | MACD Div 1.10% | Cloud 2.10%

AI Integration 🤖

The bot uses two specialized AI models:

Chat Assistance (google/gemini-flash-1.5-8b)

• High throughput (191 tokens/sec)
• 1M context window
• Multilingual support
• Cost: $0.15/1M tokens
• Used for: User interactions and queries

Technical Architecture 🏗️

Core Components

• Main Bot (main.py): Handles Discord interactions and command processing
• Command Handler (command_handler.py): Manages command parsing and execution
• Symbol Handler (symbol_handler.py): Manages real-time price data and signal generation for each symbol
• Strategy Factory (factory.py): Creates and manages trading strategy instances
• Base Strategy (base.py): Provides common interface and functionality for all strategies
• WebSocket Client (bybit_ws.py): Maintains real-time connection with Bybit exchange
• OpenRouter Client (openrouter_client.py): Handles AI chat functionality

Strategy System

The bot uses a factory pattern for strategy management:

1. Base Strategy: Abstract class defining common interface

   • Market trend analysis
   • Pattern recognition
   • Volume analysis
   • Dynamic threshold calculation

2. Strategy Factory: Central point for strategy creation

   • Manages strategy registration
   • Creates strategy instances
   • Handles strategy configuration

3. Concrete Strategies: Implement specific analysis methods

   • RSI with dynamic thresholds
   • MACD with adaptive divergence analysis
   • Bollinger Bands with squeeze detection
   • Ichimoku Cloud with crypto-optimized settings

Signal Generation Process

1. Price data is collected in real-time for all USDT perpetual pairs
2. Each strategy maintains its own DataFrame with required indicators
3. When new data arrives:
   • DataFrames are updated with the latest prices
   • Technical indicators are recalculated
   • Market conditions are analyzed:
     • Trend strength
     • Volume confirmation
     • Pattern recognition
     • Dynamic thresholds
   • Signals are generated with confidence scores
   • High-confidence signals are sent to Discord channel

AI Chat System 🤖

The bot uses OpenRouter's API to provide intelligent responses about:

• Cryptocurrency markets and trading concepts
• Bot features and strategy explanations
• Technical analysis insights

Chat features:

• Focused on crypto/trading topics
• Concise, technical responses
• No financial advice or predictions
• Automatic disclaimers for market-related info
• Response length limited to maintain channel readability

Contributing 🤝

Contributions are welcome! The bot is designed to be easily extensible, especially for adding new trading strategies.

To add a new strategy:

1. Create a new strategy class extending SignalStrategy
2. Implement required methods:
   • min_candles property
   • calculate_indicator method
   • analyze_market method
3. Add strategy type to StrategyType enum
4. Register strategy in StrategyFactory

Troubleshooting 🔍

Common issues and solutions:

• WebSocket Connection Issues: Check your internet connection and Bybit API status
• Missing Price Data: Ensure minimum candle requirements are met for your chosen strategy
• Discord Permission Errors: Verify bot has proper channel permissions
• Invalid Interval: Make sure DEFAULT_INTERVAL in .env is one of the valid values
• Strategy Errors: Check log files for specific error messages

Performance Considerations 🚀

• Efficient in-memory data structures for price data management
• Strategy-specific window sizes to optimize memory usage
• Factory pattern for efficient strategy instantiation
• WebSocket connection for real-time updates
• Batched signal processing for reduced Discord API calls

License 📄

This project is open source and available under the MIT License.

Disclaimer ⚠️

This bot is for educational purposes only. Always do your own research and never trade based solely on automated signals. Cryptocurrency trading involves substantial risk of loss.