Crypto Trading Bot

Python trading-signal bot for BTC/USDT with Discord notifications, backtesting, and parameter optimization.

This project is focused on learning and experimentation: strategy design, indicator engineering, and reproducible evaluation.

Features

• Real-time market data using CCXT (default exchange: Kraken).
• Technical indicator pipeline (EMA, RSI, ATR, MACD, VWAP, Bollinger squeeze, market structure).
• Rule-based strategy that returns BUY, SELL, or NEUTRAL signals.
• Discord notifications for signals and heartbeat status.
• Stateful signal deduplication via state.json.
• Backtest engine with fees, slippage, ATR-based stop/take-profit, partial TP, and cooldown logic.
• Optuna grid search for strategy parameter tuning.
• GitHub Actions workflow for scheduled runs.

Project Structure

Path	Purpose
main.py	Main entry point: fetch data, compute indicators, generate signal, send notifications.
src/data_fetcher.py	OHLCV/ticker retrieval via CCXT.
src/indicators.py	Indicator calculations and feature engineering.
src/strategy.py	Main strategy logic (ImprovedStrategy).
src/notifier.py	Discord webhook messaging (signal + heartbeat + test mode).
src/state_manager.py	Persistent state to avoid duplicate alerts.
test_connection.py	Manual connectivity and Discord test script.
test_new_strategy.py	Manual strategy sanity-check script.
test_simulation.py	Loop runner for local test mode.
test_backtest.py	Historical backtesting and chart output.
test_grid_search.py	Optuna optimization script.
.github/workflows/trading-bot.yml	Scheduled GitHub Actions execution.

Requirements

• Python 3.11+ recommended.
• A Discord webhook URL.
• Internet connection for exchange data.

Quick Start

1. Clone the repository.

git clone https://github.com/Willi363363/Bot-crypto.git
cd Bot-crypto

2. Create and activate a virtual environment.

python3 -m venv .venv
source .venv/bin/activate

3. Install dependencies.

pip install -r requirements.txt

4. Create your environment file.

cp .env.example .env

5. Fill at least these keys in .env:

• DISCORD_WEBHOOK_URL
• DISCORD_HEARTBEAT_WEBHOOK_URL (optional, falls back to signal webhook)

6. Run the bot.

python main.py

Main Commands

• Run live analysis once: python main.py
• Test exchange + webhook flow: python test_connection.py
• Run local loop in test mode: python test_simulation.py
• Backtest strategy: python test_backtest.py
• Run Optuna search: python test_grid_search.py

Example with custom trials:

GRID_TRIALS=100 python test_grid_search.py

Configuration

All runtime settings are environment variables. See .env.example for the complete list.

Important groups:

• Market/runtime: SYMBOL, TIMEFRAME, EXCHANGE, DATA_LIMIT, SEND_HEARTBEAT
• Discord: DISCORD_WEBHOOK_URL, DISCORD_HEARTBEAT_WEBHOOK_URL, DISCORD_TEST_WEBHOOK_URL, TEST_MODE
• Strategy filters: VOLUME_RATIO_MIN, VOLUME_SPIKE_MIN, CHOP_NO_TRADE_MAX, ATR_PCT_MIN, ATR_EXTREME_MULT, RSI_MIN, RSI_MAX
• Risk management: ATR_STOP_MULT, TP1_MULT, TP2_MULT, COOLDOWN_BARS, COOLDOWN_BARS_SL, TIME_STOP_BARS
• Backtest: INITIAL_CAPITAL, FEE_RATE, SLIPPAGE_BPS, HIST_EXCHANGE, START_DATE, WARMUP_BARS, LONG_ONLY
• Plotting: PLOT_TRADES, PLOT_PATH, PLOT_DAYS, PLOT_MAX_BARS, PLOT_LABEL_TRADES, PLOT_DEBUG, PLOT_START_DATE

GitHub Actions

The workflow in .github/workflows/trading-bot.yml runs on a schedule and can also be triggered manually.

Required GitHub repository secrets:

• DISCORD_WEBHOOK_URL
• DISCORD_HEARTBEAT_WEBHOOK_URL

Notes

• state.json is intentionally local and ignored in Git.
• data/ is ignored and used as a cache for historical OHLCV CSV files.
• This repository uses executable Python scripts for validation/backtesting rather than a full pytest suite.

Portfolio and Skills

This repository is part of my software engineering learning journey as a first-year Epitech Montpellier student.

What this project demonstrates:

• Python application architecture with modular components.
• Market data ingestion and feature engineering for time-series analysis.
• Strategy implementation and rule-based decision systems.
• Backtesting methodology with realistic constraints (fees, slippage, risk controls).
• Practical DevOps basics with scheduled GitHub Actions workflows.
• Documentation discipline and reproducible local setup.

Current focus areas:

• Writing cleaner tests and improving reliability.
• Improving strategy robustness across market regimes.
• Strengthening code quality standards for team collaboration.

Disclaimer

This software is for educational purposes only. It is not financial advice. Use at your own risk.

Contributing

See CONTRIBUTING.md.

License

MIT License. See LICENSE.