Trading Scanner — MT5 Signal Bot + Streamlit Dashboard

A professional Forex & Gold trading scanner built in Python that connects to MetaTrader 5, detects high-probability setups using Smart Money Concepts (SMC), scores them, and sends progressive Telegram alerts with chart screenshots.

Designed as Phase 1 of a full Trading Operating System — no auto-execution, full manual control.

---

Features

Market Scanner

• Connects to MetaTrader 5 via the official Python API
• Scans 6 instruments simultaneously: XAUUSD, EURUSD, GBPUSD, USDJPY, GBPJPY, EURJPY
• Multi-timeframe analysis: H4 bias → H1 structure → M15 entry
• Runs on a configurable interval (default 60s)

Strategy Logic (Smart Money Concepts)

Condition	Description
HTF Bias	Bullish/bearish trend via EMA20/EMA50 on H4
Market Structure	Swing highs/lows across timeframes
Break of Structure (BOS)	Price closes beyond last confirmed swing
Liquidity Sweep	Wick past structural level, body closes back inside
Fair Value Gap (FVG)	3-candle imbalance detection, tracks filled/unfilled
Retest	Price returning to FVG or broken structure
Rejection Candle	Hammer, shooting star, bullish/bearish engulfing
Session Filter	London, New York, London/NY overlap, Asian
Risk-to-Reward	Auto-calculated entry, SL, TP1 (2R), TP2 (3R)
News Filter	Blocks alerts 30m before / 15m after HIGH-impact events (ForexFactory)

Setup Scoring Engine

Each setup is dynamically scored (max 14 points) and graded:

Grade	Min Score	Description
🥉 Standard	4	Basic confluence
🥈 A	7	Good setup
🥇 A+	10	High-probability
💎 Elite	12	All conditions confirmed

Telegram Alerts

Progressive staged alerts sent with chart screenshots:

• 🔍 Setup Developing
• ⚡ Entry Zone Approaching
• ✅ 5-Min / 15-Min Confirmation
• 🎯 High-Confidence Setup

Each alert includes grade, score bar, bias, session, entry/SL/TP levels.

Risk Management Constitution

Automatically suspends alerts when limits are breached:

• Max % risk per trade (used for lot size calculation)
• Daily drawdown limit
• Weekly drawdown limit
• Max consecutive losses

Trade Journal & Database

SQLite database storing:

• Setups detected (symbol, grade, score, levels, chart path)
• Trades (entry, SL, TP, outcome, RR achieved, notes)
• Journal entries (daily review responses)
• Risk log (equity, drawdown history)

News Filter

• Fetches the ForexFactory weekly calendar (public JSON endpoint, no API key needed)
• Caches events locally for 1 hour, refreshes automatically
• Maps each symbol to its currencies (e.g. GBPJPY → GBP + JPY)
• Suppresses scanner alerts during the blackout window around HIGH-impact events
• Sends a Telegram notification when an alert is blocked (once per symbol, with cooldown)
• Window is fully configurable via .env

Streamlit Dashboard

Seven-page dashboard:

Page	Contents
Dashboard	Today's stats, recent setups, next high-impact event countdown, equity trend
Economic Calendar	Live ForexFactory events, countdown timer, impact filter, timeline chart, scanner blackout indicator
Active Setups	Filterable setup table with chart previews
Trade Journal	Log and close trades
Analytics	Win rate, RR distribution, cumulative RR curve, performance by grade/symbol/session
Daily Review	End-of-day reflection form (stored for future AI coaching)
Risk Status	Drawdown tracker, consecutive loss chart, constitution rules

---

Project Structure

trading-scanner/
├── main.py                    # Scanner entry point (runs the loop)
├── config.py                  # All settings loaded from .env
├── scanner/
│   ├── mt5_connector.py       # MT5 connection and bar fetching
│   ├── market_analyzer.py     # Full SMC strategy logic
│   ├── scoring_engine.py      # Setup scoring and grading
│   ├── chart_generator.py     # Dark-theme candlestick charts (mplfinance)
│   └── news_filter.py         # ForexFactory calendar + blackout logic
├── alerts/
│   └── telegram_bot.py        # Telegram message and photo sending
├── risk/
│   └── risk_manager.py        # Constitution rules enforcement + lot sizing
├── database/
│   └── db_manager.py          # SQLite schema + all CRUD operations
├── dashboard/
│   └── app.py                 # Streamlit 7-page dashboard
├── requirements.txt
└── .env.example

---

Setup

Requirements

• Windows (MetaTrader5 Python package is Windows-only)
• MetaTrader 5 terminal installed and logged in
• Python 3.10+
• A Telegram bot (create via @BotFather)

Install

git clone https://github.com/NadirAliOfficial/trading-scanner.git
cd trading-scanner
pip install -r requirements.txt

Configure

cp .env.example .env

Edit .env:

TELEGRAM_BOT_TOKEN=your_bot_token
TELEGRAM_CHAT_ID=your_chat_id
MT5_LOGIN=123456
MT5_PASSWORD=your_password
MT5_SERVER=YourBroker-Server

Run the Scanner

python main.py

Run the Dashboard

streamlit run dashboard/app.py

---

Configuration Reference

Variable	Default	Description
SCAN_INTERVAL	60	Seconds between scans
ALERT_COOLDOWN_MINUTES	60	Min minutes between repeat alerts per symbol/grade
MAX_RISK_PER_TRADE	1.0	% of account per trade
MAX_DAILY_DRAWDOWN	3.0	Daily DD % limit
MAX_WEEKLY_DRAWDOWN	6.0	Weekly DD % limit
MAX_CONSECUTIVE_LOSSES	3	Losses before scanner pauses
SCORE_STANDARD	4	Min score for Standard alert
SCORE_A	7	Min score for A alert
SCORE_A_PLUS	10	Min score for A+ alert
SCORE_ELITE	12	Min score for Elite alert
NEWS_FILTER_ENABLED	true	Enable/disable news blackout
NEWS_BLOCK_MINUTES_BEFORE	30	Minutes before event to block alerts
NEWS_BLOCK_MINUTES_AFTER	15	Minutes after event to resume alerts

---

Roadmap

• AI coaching feedback based on journal entries
• Multi-broker support
• Webhook-based Telegram bot for journal commands
• Machine learning setup quality predictor

---

Disclaimer

This tool is for decision support only. It does not place trades automatically. All execution is manual. Past performance of any strategy does not guarantee future results. Trade responsibly.

---

License

MIT