Autonomous Silver Trading Bot

Version: 1.0
Created: February 7, 2026
Platform: Interactive Brokers (IBKR) via TWS API
Language: Python 3.11+
Status: Paper trading ready, live trading requires approval

---

Overview

A fully autonomous trading bot designed to trade SILVER (SLV ETF and silver futures) with optional cryptocurrency trading during off-market hours. Built specifically for McKinzie's trading account with strict risk management and Telegram reporting.

Key Features

✅ Multiple Strategies:

• Mean reversion (RSI + Bollinger Bands)
• Grid trading (range-bound markets)
• Automatic strategy selection based on volatility

✅ Risk Management:

• 2% per-trade risk limit
• 5% daily loss limit (auto-stop)
• 15% total drawdown limit (emergency stop)
• Position size calculator
• Trailing stop losses

✅ Telegram Integration:

• Real-time trade alerts
• Daily performance summaries
• Risk limit warnings
• Connection status updates

✅ 24/7 Trading:

• Silver during market hours (7:30 AM - 2:00 PM MT)
• Crypto (ETH/BTC) during off-hours (optional)

✅ Complete Logging:

• SQLite database for all trades
• Performance analytics
• Equity curve tracking
• Strategy comparison

✅ Backtesting:

• Test strategies on historical data
• Compare performance metrics
• Risk-free strategy validation

---

Quick Start

1. Setup

cd /Users/mmcassistant/clawd/projects/autonomous-trader
./setup.sh

This will:

• Create Python virtual environment
• Install all dependencies
• Initialize database
• Test connections

2. Configure Environment

Edit ~/.clawdbot/.env:

# Telegram (required)
TELEGRAM_BOT_TOKEN=your_bot_token
TELEGRAM_CHAT_ID=8417770794

# For live trading (keep false for paper)
IBKR_LIVE_APPROVED=false

# Crypto (optional, for off-hours trading)
CRYPTO_API_KEY=your_kraken_api_key
CRYPTO_API_SECRET=your_kraken_secret

3. Start Interactive Brokers

Paper Trading:

1. Open TWS or IB Gateway
2. Login with paper trading account
3. Go to: File → Global Configuration → API → Settings
4. Enable "ActiveX and Socket Clients" ✓
5. Socket port: 4002 (paper) or 4001 (live)
6. Add trusted IP: 127.0.0.1
7. Click OK and restart TWS

4. Run Backtest (Recommended First)

source venv/bin/activate
python3 backtest.py

This downloads recent SLV data and tests the mean reversion strategy. Review results before live trading.

5. Start Bot

source venv/bin/activate
python3 bot.py

Bot runs every 15 minutes, analyzing market data and executing trades when signals trigger.

To stop: Press Ctrl+C for graceful shutdown

---

Trading Strategies

Strategy 1: Mean Reversion (Primary)

Entry Signals (BUY):

1. Strong: RSI < 30 AND price at lower Bollinger Band → 90% confidence
2. Panic Buy: Daily drop ≥ 3% → 85% confidence (catch panic selling)
3. Moderate: RSI < 30 only → 60% confidence

Exit Signals (SELL):

1. RSI > 70 (overbought)
2. Price hits upper Bollinger Band
3. Stop loss hit (2% below entry)
4. Take profit hit (4% above entry)

Best For: High volatility periods (5-10% daily swings)

Example Trade:

SLV drops from $80 to $73 (-9%) in one day
RSI: 25 (oversold)
Price: $73.20 (at lower BB)
→ BUY signal: 90% confidence
→ Enter: $73.20, Stop: $71.74, Target: $76.13
→ Risk/Reward: 2:1

Strategy 2: Grid Trading (Secondary)

Setup:

• Create 5 buy levels at 2% intervals below current price
• Create 5 sell levels at 2% intervals above current price
• Execute limit orders at each level

Example Grid at $78:

SELL Levels:
$79.56 (+2.0%)  ←─┐
$79.17 (+1.5%)    │
$78.78 (+1.0%)    │ Take profits
$78.39 (+0.5%)    │
                  │
$78.00 ←─────────┬─ Current price
                  │
$77.61 (-0.5%)    │
$77.22 (-1.0%)    │ Buy opportunities
$76.83 (-1.5%)    │
$76.44 (-2.0%)  ←─┘

Best For: Range-bound, lower volatility markets

Grid resets when:

• Price moves >10% from grid center
• Weekly (every Sunday)

Strategy 3: Crypto Off-Hours (Optional)

ETH Grid Trading:

• 1% spacing (tighter than silver due to volatility)
• Active during non-market hours
• Auto-pauses during market hours

BTC Mean Reversion:

• 4-hour timeframe
• RSI (21 period)
• Less aggressive than silver

When Active:

• Weekends
• Weekday pre-market (before 7:30 AM MT)
• Weekday after-hours (after 2:00 PM MT)

---

Risk Management

Position Sizing

Formula:

Position Size = min(
    Account * 10%,  # Base size
    (Account * 2%) / (Entry - Stop)  # Risk-based
)

Limits:

• Never exceed 30% of account in one position
• Maximum 3 concurrent positions
• Positions auto-reduced by 50% if weekly loss > 10%

Stop Losses

Types:

1. Fixed: 2% below entry (default)
2. ATR-based: 2x Average True Range (for volatile markets)
3. Trailing: 1.5% trailing stop (locks in profits)

Example:

Buy SLV @ $78.00
Fixed stop: $76.44 (-2%)
Take profit: $81.12 (+4%)
If price → $81, trailing stop → $79.79

Risk Limits (Enforced Automatically)

Limit	Threshold	Action
Per Trade	2% of account	Max risk per position
Daily Loss	5% of account	⚠️ STOP ALL TRADING for the day
Weekly Loss	10% of account	⚠️ Reduce position sizes 50%
Total Drawdown	15% from peak	🚨 STOP ALL TRADING + alert McKinzie

When limits hit:

• Automatic halt
• Telegram alert sent
• Manual intervention required to resume
• All positions evaluated

---

Telegram Notifications

Trade Alerts

🟢 BUY EXECUTED

📊 SLV
Quantity: 12 shares
Price: $78.45
Total: $941.40

Strategy: silver_mean_reversion
Reason: RSI oversold (28.5) + at lower BB

⏰ 2026-02-07 10:30:15 MT

Daily Summary (4 PM MT)

📈 DAILY TRADING SUMMARY
2026-02-07

💵 Account
Value: $103.50
Daily P&L: +$3.50 (+3.5%)

📊 Trading Activity
Trades: 5 (3W/2L)
Win Rate: 60.0%

🎯 Open Positions
• SLV: 10 shares @ $78.50 ($785.00)

📈 Strategy Performance
• silver_mean_reversion: 3 trades, +2.5%
• silver_grid: 2 trades, +1.0%

Risk Alerts

🚨 RISK ALERT - CRITICAL

Type: TOTAL_DRAWDOWN_LIMIT

Drawdown reached 15.2% (limit: 15.0%)
ALL TRADING STOPPED

Current Metrics:
• Account: $84.80
• Peak: $100.00
• Drawdown: $15.20 (-15.2%)

⏰ 2026-02-07 11:45:23 MT

---

Database & Logging

Trade Log (trades.db)

Schema:

trades:
  - id, timestamp, symbol, action
  - quantity, price, total_value
  - strategy, reason
  - pnl, pnl_pct
  - entry_price, exit_price, hold_duration

account_snapshots:
  - timestamp, account_value, cash
  - positions_value, daily_pnl, total_return_pct

positions:
  - symbol, quantity, entry_price, entry_time
  - strategy, stop_loss, take_profit

Query Trades

from trade_logger import TradeLogger

tl = TradeLogger()

# Get today's trades
today_trades = tl.get_daily_trades()

# Performance metrics
metrics = tl.calculate_performance_metrics()
print(f"Win rate: {metrics['win_rate']:.1f}%")

# Generate report
print(tl.generate_report('today'))
print(tl.generate_report('week'))
print(tl.generate_report('all'))

Performance Metrics

Tracked automatically:

• Win Rate: % of profitable trades
• Profit Factor: Total wins / total losses
• Sharpe Ratio: Risk-adjusted returns
• Max Drawdown: Largest peak-to-trough decline
• Average Win/Loss: Mean P&L per trade
• Best/Worst Trade: Extremes

---

Backtesting

Run Backtest

python3 backtest.py

Downloads last 3 months of SLV data and simulates strategy.

Output:

============================================================
BACKTEST RESULTS
============================================================

Final Capital: $112.50
Total Return: +$12.50 (+12.50%)

Total Trades: 45
Win Rate: 62.2%
Average Win: $1.85
Average Loss: $0.92
Profit Factor: 2.15

Sharpe Ratio: 1.82
Max Drawdown: 8.34%
============================================================

Custom Backtest

from backtest import Backtest

bt = Backtest(initial_capital=100.0)

# Download specific date range
df = bt.download_slv_data(
    start_date="2025-01-01",
    end_date="2026-02-01"
)

# Test strategy
metrics = bt.run_strategy(df, strategy_name="mean_reversion")

# Plot results
bt.plot_results(save_path="my_backtest.png")

Compare Strategies

from backtest import compare_strategies

compare_strategies(df)

Runs multiple strategies on same data and outputs comparison table.

---

Configuration

Main Settings (config.py)

Capital & Risk:

STARTING_CAPITAL = 100.0
MAX_RISK_PER_TRADE_PCT = 0.02  # 2%
DAILY_LOSS_LIMIT_PCT = 0.05     # 5%
TOTAL_DRAWDOWN_LIMIT_PCT = 0.15 # 15%

Position Sizing:

MAX_POSITION_SIZE_PCT = 0.30  # 30% max in one position
POSITION_SIZE_PER_TRADE_PCT = 0.10  # 10% base size

Strategy Parameters:

RSI_OVERSOLD = 30
RSI_OVERBOUGHT = 70
BB_PERIOD = 20
BB_STD_DEV = 2.0
GRID_SPACING_PCT = 0.02  # 2%

Stop Loss & Take Profit:

STOP_LOSS_PCT = 0.02      # 2%
TAKE_PROFIT_PCT = 0.04    # 4%
TRAILING_STOP_PCT = 0.015 # 1.5%

Validate Config

python3 config.py

Checks all settings are valid before trading.

---

File Structure

autonomous-trader/
├── bot.py                  # Main orchestrator
├── config.py               # All configuration
├── indicators.py           # Technical indicators (RSI, BB, MACD, etc.)
├── strategy_silver.py      # Silver strategies
├── strategy_crypto.py      # Crypto strategies
├── risk_manager.py         # Risk limits & position sizing
├── trade_logger.py         # Database & logging
├── reporter.py             # Telegram notifications
├── backtest.py             # Backtesting engine
├── requirements.txt        # Python dependencies
├── setup.sh                # Setup script
├── SECURITY.md             # Security rules
├── trades.db               # SQLite database
└── backups/                # Daily database backups

---

OpenClaw Integration

Cron Job Setup

Add to OpenClaw cron:

# Run trading bot every 15 minutes during market hours
*/15 7-14 * * 1-5 cd /Users/mmcassistant/clawd/projects/autonomous-trader && source venv/bin/activate && python3 bot.py --run-once

Agent Skills

The agent can:

• Start/stop the bot
• Check performance
• Generate reports
• Adjust risk parameters
• Review trade logs
• Send manual Telegram alerts

Example commands:

# Check today's performance
exec("cd /projects/autonomous-trader && source venv/bin/activate && python3 -c 'from trade_logger import TradeLogger; tl = TradeLogger(); print(tl.generate_report(\"today\"))'")

# Get risk status
exec("cd /projects/autonomous-trader && source venv/bin/activate && python3 -c 'from risk_manager import RiskManager; rm = RiskManager(100); print(rm.get_risk_summary())'")

---

Silver Market Context

Current Status (Feb 2026)

• Price: ~$78/oz (SLV ~$78/share)
• YoY Return: +145% (massive bull run)
• Volatility: 5-10% daily swings
• Recent: Dropped 10%, bounced 4% same day

Why Silver Now?

✅ High volatility = opportunity for mean reversion
✅ Strong trend but frequent pullbacks
✅ Liquid market (SLV has high volume)
✅ Clear support/resistance levels

Trading Hours

Silver (SLV):

• NYSE: 9:30 AM - 4:00 PM ET (7:30 AM - 2:00 PM MT)
• Pre-market: 4:00 AM - 9:30 AM ET
• After-hours: 4:00 PM - 8:00 PM ET

Silver Futures (SI):

• COMEX: Nearly 24 hours (Sunday 6 PM - Friday 5 PM ET)

---

Troubleshooting

Bot won't connect to IBKR

Check:

1. Is TWS/Gateway running?
2. API enabled? (File → Global Config → API → Settings)
3. Port correct? (4002 for paper, 4001 for live)
4. Trusted IP added? (127.0.0.1)
5. TWS restarted after config change?

Test connection:

python3 -c "
import asyncio
from ib_insync import IB
async def test():
    ib = IB()
    await ib.connectAsync('127.0.0.1', 4002, clientId=999, timeout=5)
    print('Connected!')
    ib.disconnect()
asyncio.run(test())
"

No Telegram alerts

Check:

1. Bot token set in ~/.clawdbot/.env?
2. Chat ID correct? (8417770794)
3. Bot added to chat?

Test:

python3 reporter.py

Bot not trading

Possible reasons:

1. Risk limit hit (check get_risk_summary())
2. Insufficient signal strength (<70%)
3. No clear signals in current market
4. Daily loss limit reached
5. Paper mode vs live mode confusion

Debug:

python3 -c "
from config import validate_config, get_config_summary
if validate_config():
    print('Config valid')
    print(get_config_summary())
"

Database errors

Backup and recreate:

cp trades.db backups/trades_$(date +%Y%m%d).db
python3 -c "from trade_logger import TradeLogger; TradeLogger()"

---

Safety Checklist

Before going live:

• Tested in paper mode for 30+ days
• Strategies are profitable
• Risk limits never breached
• All Telegram alerts working
• Database logging working
• Emergency stop tested
• Reviewed all trades manually
• Understand all strategies
• Read SECURITY.md completely
• McKinzie approval obtained
• Set IBKR_LIVE_APPROVED=true in .env
• Changed PAPER_TRADING=False in config.py

Start small: Even in live mode, keep capital at $100 initially.

---

Performance Expectations

Conservative Estimates

Based on backtests and silver volatility:

Monthly:

• Target: 5-10% return
• Max drawdown: 8-12%
• Win rate: 55-65%
• Sharpe ratio: 1.5+

Yearly:

• Target: 60-120% return (if silver trend continues)
• Max drawdown: 15% (hard stop)

Reality check:

• Past performance ≠ future results
• Silver is volatile (good and bad)
• Risk management is key
• Stay disciplined

---

Support

Questions? Ask McKinzie or check:

• Project files: /Users/mmcassistant/clawd/projects/autonomous-trader/
• Security rules: SECURITY.md
• Code comments: Each file heavily documented

Issues?

1. Check logs
2. Review trade history
3. Test in paper mode
4. Contact McKinzie via Telegram

---

Future Enhancements

Planned features:

• ✨ Crypto exchange integration (Kraken API)
• ✨ Machine learning signal strength
• ✨ Multi-symbol correlation analysis
• ✨ Advanced chart pattern recognition
• ✨ Sentiment analysis (Twitter, news)
• ✨ Options strategies (covered calls)
• ✨ Portfolio rebalancing
• ✨ Tax loss harvesting

---

Version History

v1.0 (Feb 7, 2026):

• Initial release
• Mean reversion strategy
• Grid trading strategy
• Risk management
• Telegram integration
• Backtesting engine
• Paper trading ready

---

Built with ❤️ for McKinzie's trading success.

Trade smart. Manage risk. Let the bot work for you.