Zion Smart DCA — BTC Accumulation Strategy Skill

An intelligent DCA strategy that dynamically adjusts BTC purchase amounts
based on CMC Fear & Greed Index, RSI signals, and halving cycle analysis,
with a built-in reserve that auto-deploys during market capitulation events.

v4.0 features: Cycle reading (Pillar 0), double-layer Buildup confirmation,
flexible frequency (daily/weekly/biweekly/monthly), Black Swan Protocol, and fiscal awareness.

Proven over 5 years of backtesting: +56.4% vs standard DCA | Sharpe 1.768 | Max DD -44.6%
57 real transactions since February 2026 — this strategy is live, not theoretical.

Prerequisites

• Python 3.8+
• CMC API Key: https://pro.coinmarketcap.com/login
• pip install pandas numpy requests python-dotenv yfinance bnbagent
• CMC MCP Server configured (optional — skill works with REST API fallback)

CMC Tools Used

• get_global_metrics_latest — Fear & Greed Index (primary buy signal)
• get_crypto_technical_analysis — RSI 14-period for BTC (Buildup trigger)
• get_crypto_quotes_latest — BTC current price
• get_upcoming_macro_events — macro context (Fed decisions, CPI)
• trending_crypto_narratives — market sentiment context

Usage

# Live decision — default $100 weekly
python src/zion_dca_skill.py --budget 100

# Flexible frequency — adapts to your reality
python src/zion_dca_skill.py --budget 10 --frequency daily
python src/zion_dca_skill.py --budget 100 --frequency weekly
python src/zion_dca_skill.py --budget 200 --frequency biweekly
python src/zion_dca_skill.py --budget 500 --frequency monthly

# Run full 5-year backtest
python backtest/backtest.py --symbol BTC --days 1825 --budget 100

# Run with custom parameters
python backtest/backtest.py \
  --budget 100 \
  --rsi-threshold 35 \
  --fg-extreme-fear 24 \
  --fg-extreme-greed 75

Parameters

Output / Performance Metrics (5-Year Backtest 2021–2026)

Decision Output Format

{
  "skill": "Zion Smart DCA",
  "version": "4.0",
  "decision": {
    "action": "BUY",
    "type": "BUILDUP_2X",
    "multiplier": 2.0,
    "amount_usd": 140.00,
    "btc_amount": 0.00231405,
    "reserve_contribution_usd": 30.00
  },
  "market": {
    "btc_price_usd": 60500.00,
    "fear_greed_index": 11,
    "rsi_14d": 33.5,
    "rsi_weekly": 42.0
  },
  "pillars": {
    "pillar_0_cycle": {
      "phase": "Accumulation",
      "months_since_halving": 8,
      "adjustment": "aggressive"
    },
    "pillar_2_buildup": {
      "active": true,
      "context": "200WMA confirmed",
      "max_slots": 5,
      "consecutive_days": 1
    }
  },
  "reasoning": [
    "Cycle Phase: Accumulation (8mo post-halving) → aggressive",
    "Extreme Fear (F&G=11 <= 24) → 2.0x — maximum accumulation",
    "RSI 33.5 < 35.0 → Buildup ELIGIBLE (oversold + context confirmed)",
    "Reserve First: 70% DCA + 30% reserve"
  ],
  "rules_applied": ["Rule 1", "Rule 2", "Rule 3", "Rule 4", "Rule 5", "Rule 8"]
}

F&G Multiplier Table (v4.0)

Risks & Caveats

• Past performance does not guarantee future results
• Strategy assumes sufficient capital reserve for Buildup events (min 2x base budget)
• RSI and F&G are lagging indicators — some signals may arrive 1-3 days late
• Reserve deployment requires manual execution (not automated in Track 2 submission)
• BTC-focused strategy — altcoin allocation rules apply separately
• Backtest uses weekly granularity; intra-week volatility not captured

Strategy Rules Summary (v4.0)

BNB AI Agent SDK Integration

This skill registers an on-chain identity via ERC-8004 standard using the BNB AI Agent SDK,
enabling discovery, reputation tracking, and future integration with the BNB Chain ecosystem.

from bnbagent import AgentIdentity
agent = AgentIdentity(name="Zion Smart DCA", version="4.0")
agent.register()  # ERC-8004 on BNB testnet

License

MIT — See LICENSE

BNB Hack 2026 | Track 2: Strategy Skills
CoinMarketCap × BNB Chain × Trust Wallet