🚀 SignalEngine v2 — SLC-Only Crypto Futures Trading Bot

An autonomous crypto futures trading bot powered by the SLC Execution Blueprint (Structure → Level → Confirmation). Monitors Binance Futures via WebSocket, delivering real-time signals to Discord with self-improving ML filtering.

Architecture

┌──────────────────────────────────────────────────┐
│                   index.ts                        │
│         (Binance Engine Bootstrap)                │
├──────────────────────────────────────────────────┤
│              BinanceProvider (WebSocket)           │
│              ↓ Zero-delay candle close             │
├──────────────────────────────────────────────────┤
│                 SignalEngine v2                    │
│  ┌──────────────────────────────────────────┐    │
│  │  SLC Strategy (ONLY strategy)            │    │
│  │     ├─ Stage 1: 4H Structure (EMA 21/50) │    │
│  │     ├─ Stage 2: 15m Supply/Demand Zones  │    │
│  │     └─ Stage 3: Stochastic (5,3,3)       │    │
│  ├──────────────────────────────────────────┤    │
│  │  ML Filter (Multi-Dimensional Brain)      │    │
│  │     ├─ Pattern win-rate (base)            │    │
│  │     ├─ Volume confidence (+/- bonus)      │    │
│  │     ├─ Session performance (time-of-day)  │    │
│  │     └─ Loss streak protection             │    │
│  └──────────────────────────────────────────┘    │
│            ↓                                      │
│  TradeTracker (SL/TP/Breakeven/Trailing/MFE/MAE)  │
│            ↓                                      │
│  ┌─ DatabaseManager (SQLite) ─────────────────┐  │
│  │  active_trades → crash recovery             │  │
│  │  virtual_trades → ML learning data          │  │
│  └─────────────────────────────────────────────┘  │
│            ↓                                      │
│  Discord Notifier (Signals / Trades / Results)    │
│            ↓                                      │
│  OrderExecutor (Binance Testnet Demo Trading)     │
└──────────────────────────────────────────────────┘

Features

SLC-Only Strategy — 3-stage mechanical system (Structure → Level → Confirmation). No fallback strategies.
50 Symbols — Monitors top 50 Binance Futures pairs by volume
Multi-Timeframe — 4H macro trend + 15m precision entries
Volume-Weighted Zones — Supply/Demand scored by institutional volume (🔥 = 1.5x+ avg)
Crash Recovery — Active trades persist in SQLite; recovers on restart
Self-Improving ML — 4-factor scoring: pattern win rate, volume confidence, session performance, streak protection
MFE/MAE Tracking — Records best and worst unrealized PnL per trade (answers "should I take partial profits?")
Demo Trading — Real orders on Binance Testnet (fake USDT, zero risk)
ATR Trailing Stops — Auto-tightens SL using 1.5x ATR after breakeven
Graceful Shutdown — SIGINT/SIGTERM saves state before exit
Exponential Reconnect — WebSocket reconnection with backoff (5s → 60s max)

Tech Stack

Component	Technology
Language	TypeScript (ESNext)
Runtime	Node.js 20+
Database	SQLite (WAL mode)
Indicators	`technicalindicators` (EMA, ADX, ATR, Stochastic)
WebSocket	`ws` library
HTTP	`axios`
Notifications	Discord Webhooks

Project Structure

src/
├── index.ts                    # Entry point — Binance-only bootstrap
├── backtest.ts                 # CLI backtesting entry point
├── engines/
│   ├── SignalEngine.ts         # Core SLC orchestration + trade lifecycle
│   └── BacktestEngine.ts       # Historical data replay + statistics
├── strategies/
│   └── SLCStrategy.ts          # SLC Blueprint (Structure/Level/Confirmation)
├── providers/
│   ├── MarketProvider.ts       # OHLCV interface
│   └── BinanceProvider.ts      # Binance Futures WebSocket + REST
├── services/
│   ├── MLAnalyzer.ts           # Multi-dimensional ML signal filter
│   ├── DatabaseManager.ts      # SQLite (trade history + crash recovery)
│   ├── VirtualTradeTracker.ts  # SL/TP/Trailing/MFE/MAE management
│   ├── OrderExecutor.ts        # Binance testnet order placement
│   ├── SymbolService.ts        # Top-50 volume symbol discovery
│   ├── CurrencyService.ts      # USDT formatting
│   └── DiscordNotifier.ts      # Discord webhook sender

Setup

1. Install Dependencies

npm install

2. Configure Environment

Create a .env file:

# Binance Futures API
BINANCE_API_KEY=your_testnet_api_key
BINANCE_SECRET_KEY=your_testnet_secret_key
BINANCE_TESTNET=true                  # Use testnet (demo)

# Demo Trading
ENABLE_TESTNET_ORDERS=true            # Place real orders with fake money

# Symbol Count (default: 50)
SYMBOL_COUNT=50

# Discord Webhooks
BINANCE_SIGNALS_WEBHOOK_URL=https://discord.com/api/webhooks/...
BINANCE_TRADES_WEBHOOK_URL=https://discord.com/api/webhooks/...
BINANCE_RESULTS_WEBHOOK_URL=https://discord.com/api/webhooks/...

3. Build & Run

npx tsc && node dist/index.js

Backtesting

Validate the SLC strategy against historical data before trusting signals.

Single Symbol

npx tsc && node dist/backtest.js BTCUSDT 3    # 3-month backtest
node dist/backtest.js ETHUSDT 6               # 6-month backtest

Multi-Symbol Batch

node dist/backtest.js --multi 3    # Test 8 symbols over 3 months

Reading Results

=======================================================
  📊 BACKTEST REPORT: BTCUSDT
=======================================================
  Total Trades:          12
  Wins / Losses / BE:    7 / 4 / 1
  Win Rate:              58.3%      ← Above 55% is good
  Total PnL:             +4.21%
  Avg Win:               +0.89%
  Avg Loss:              -0.42%
  Profit Factor:         3.71       ← Above 1.5 is profitable
  Max Drawdown:          -1.23%     ← How much you'd lose at worst
  Max Consec. Losses:    2
=======================================================

Metric	Good	Bad
Win Rate	>55%	<45%
Profit Factor	>1.5	<1.0
Max Drawdown	<5%	>10%
Avg Win / Avg Loss	>1.5	<1.0

Note: SLC is a sniper strategy. Expect 0-3 signals per day across 50 symbols. This is by design — fewer trades = higher conviction.

Demo Trading (Testnet)

1. Get Testnet API Keys

Go to Binance Futures Testnet
Log in with GitHub
Go to API Management → Create API key
Copy the API Key and Secret Key

2. Enable in `.env`

BINANCE_API_KEY=your_testnet_api_key
BINANCE_SECRET_KEY=your_testnet_secret_key
BINANCE_TESTNET=true
ENABLE_TESTNET_ORDERS=true

3. Run

npx tsc && node dist/index.js
# Output: 🚀 SignalEngine v2 Online
#         [Config] Orders: ✅ Demo Trading ENABLED

Safety Architecture

ENABLE_TESTNET_ORDERS=true + BINANCE_TESTNET=true
→ ✅ Demo orders with fake USDT (safe)

ENABLE_TESTNET_ORDERS=true + BINANCE_TESTNET=false  
→ 🛑 BLOCKED (requires ENABLE_LIVE_ORDERS=true)

ENABLE_TESTNET_ORDERS not set
→ Signals only, no orders placed

How the Bot Learns

The bot uses a self-improving feedback loop with 4-factor ML scoring:

Phase 1: Data Collection (First 1-3 Days)

ML Filter approves all signals (needs 20+ trades per pattern)
Every trade is logged with rich metadata: zone strength, volume score, stochastic values, session, MFE/MAE
Let it run undisturbed — it's building its knowledge base

Phase 2: Pattern Learning (After ~100 Trades)

The ML brain starts making decisions based on:

Factor	What It Does
Pattern Win Rate	Base approval check (>50%)
Volume Confidence	Zones with 1.5x+ avg volume get a 5% threshold reduction (more permissive)
Session Performance	Auto-penalizes time-of-day windows with historically bad results
Streak Protection	After 3+ consecutive losses, temporarily raises approval threshold

Phase 3: Continuous Improvement

Every trade result updates the ML brain
Bad patterns get filtered out automatically
Good sessions get rewarded
The approval threshold dynamically adjusts per signal

Training Tips

Run 24/7 for at least 1 week before trusting results
Never delete trade_history.db — it's the bot's memory
Check maxFavorable column — if trades consistently hit +1.5R before reverting, consider adding partial TP

Query the database to see ML insights:

sqlite3 trade_history.db "SELECT timeOfDay, COUNT(*), 
  ROUND(AVG(CASE WHEN outcome='WIN' THEN 1.0 ELSE 0.0 END)*100,1) as win_rate 
  FROM virtual_trades GROUP BY timeOfDay"

Crash Recovery

If the bot crashes or restarts while trades are active:

Active trades are persisted to SQLite (active_trades table)
On restart, the bot loads them back into memory
SL/TP/trailing management resumes immediately
Discord notification: "🔄 Recovered N active trades from last session"

Note: If price blew through SL/TP while the bot was offline, the trade closes at the first available price. For demo trading, Binance testnet SL/TP orders handle this automatically.

Hosting

Recommended: VPS (Virtual Private Server)

The bot needs 24/7 uptime with stable internet for WebSocket connections.

Spec	Minimum	Recommended
CPU	1 vCPU	2 vCPU
RAM	512 MB	1 GB
Storage	1 GB SSD	5 GB SSD
OS	Ubuntu 22.04+	Ubuntu 24.04 LTS
Node.js	v20+	v22 LTS

VPS Providers

Provider	Plan	~Cost/Month
Hetzner (Best Value)	CX22	~€4/mo (~₹380)
Oracle Cloud	Free Tier (ARM)	Free forever
Contabo	VPS S	~€6/mo (~₹570)
DigitalOcean	Basic Droplet	$6/mo (~₹500)

VPS Deploy

# 1. SSH into VPS
ssh root@your-vps-ip

# 2. Install Node.js
curl -fsSL https://deb.nodesource.com/setup_22.x | sudo -E bash -
sudo apt install -y nodejs

# 3. Clone & Setup
git clone https://github.com/your-user/SignalEngine.git
cd SignalEngine && npm install && npx tsc

# 4. Configure
cp .env.example .env && nano .env

# 5. Run with PM2 (auto-restart)
npm install -g pm2
pm2 start dist/index.js --name signal-engine
pm2 save && pm2 startup

Environment Variables

Variable	Required	Description
`BINANCE_API_KEY`	✅	Binance Futures API key (testnet or mainnet)
`BINANCE_SECRET_KEY`	✅	Binance Futures secret
`BINANCE_TESTNET`	❌	`true` for demo trading (default: false)
`ENABLE_TESTNET_ORDERS`	❌	`true` to place orders on testnet
`ENABLE_LIVE_ORDERS`	❌	`true` to unlock mainnet (DANGER)
`SYMBOL_COUNT`	❌	Number of top-volume symbols to monitor (default: 50)
`BINANCE_SIGNALS_WEBHOOK_URL`	❌	Discord signals channel
`BINANCE_TRADES_WEBHOOK_URL`	❌	Discord trades channel
`BINANCE_RESULTS_WEBHOOK_URL`	❌	Discord results channel

Discord Channels

Channel	Purpose	Examples
#signals	New SLC signals	`🎯 SLC SIGNAL — BTC/USDT` with checklist, market data
#trades	Trade lifecycle	`🎯 TRADE OPENED`, `🛡️ BREAKEVEN`, `📈 TRAILING`
#results	Outcomes	`🏆 WIN (+1.82%)`, `🛑 LOSS (-0.45%)`, `🔄 RECOVERED`

Disclaimer

USE AT YOUR OWN RISK. This software is provided "as is" and for educational/demonstration purposes only. The creator(s) and contributor(s) of this project assume no responsibility or liability for any financial losses, damages, or unintended consequences resulting from the use of this software. By using this bot, you acknowledge that cryptocurrency trading carries significant risk and you are solely responsible for any trades or decisions made.

License

This project is licensed under the MIT License.

Name		Name	Last commit message	Last commit date
Latest commit History 50 Commits
src		src
.env.example		.env.example
.gitignore		.gitignore
LICENSE.md		LICENSE.md
README.md		README.md
package.json		package.json
tsconfig.json		tsconfig.json

Folders and files

Latest commit

History

Repository files navigation

🚀 SignalEngine v2 — SLC-Only Crypto Futures Trading Bot

Architecture

Features

Tech Stack

Project Structure

Setup

1. Install Dependencies

2. Configure Environment

3. Build & Run

Backtesting

Single Symbol

Multi-Symbol Batch

Reading Results

Demo Trading (Testnet)

1. Get Testnet API Keys

2. Enable in .env

3. Run

Safety Architecture

How the Bot Learns

Phase 1: Data Collection (First 1-3 Days)

Phase 2: Pattern Learning (After ~100 Trades)

Phase 3: Continuous Improvement

Training Tips

Crash Recovery

Hosting

Recommended: VPS (Virtual Private Server)

VPS Providers

VPS Deploy

Environment Variables

Discord Channels

Disclaimer

License

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

2. Enable in `.env`

Packages