Provides a strategy skill for the CoinMarketCap Agent Hub, enabling agents to fetch BTC market regime and risk posture via the MCP server.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Market Regime Oracle what is the current BTC market regime and posture?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A 5-signal ā 5-state BTC market-regime classifier with posture mapping, backtested vs buy-and-hold.
Fuses momentum, sentiment, volatility, funding & flow into one explainable regime ā and a documented risk posture per regime.
Ask "What kind of BTC market is this, and how much risk should I take?"
The oracle answers with one of 5 regimes + a documented posture:
Regime | Target Exposure | Posture |
š¢ | 100% | Uptrend ā full exposure |
š” | 40% | Sideways ā light exposure |
šµ | 20% | Downtrend ā defensive |
š“ | 10% | Panic ā max defensive |
š£ | 30% | Blow-off ā take profit |
Built as an MCP Strategy Skill for the CoinMarketCap Agent Hub (BNB AI Trading ā Track 2). Any MCP-compatible client (Claude Desktop, Cursor, CMC Agent Hub) calls get_market_regime to get a deterministic, no-look-ahead risk posture.
Related MCP server: PreReason-mcp
In a down year for BTC (ā37%), the regime strategy did ā12.7% ā halving drawdown (ā24% vs ā51%) and halving volatility (19% vs 43%), outperforming buy-and-hold by ~25 points while still going long in uptrends.
Metric | Regime Strategy | Buy & Hold |
Total return | ā12.7% | ā37.4% |
Max drawdown | ā24.2% | ā51.2% |
Annualized volatility | 19.3% | 43.1% |
Sharpe (rf 4%) | ā0.82 | ā0.97 |
Sortino | ā1.01 | ā1.32 |
Data: CoinGecko BTC daily (2025-06-19 ā 2026-06-17, 364 days). Start $10,000. 10 bps/turnover cost.
CoinGecko (BTC OHLCV) alternative.me (Fear & Greed)
ā ā
āāāāāāāāāāāāāāāā¬āāāāāāāāāāāāāāāā
ā¼
āāāāāāāāāāāāāāāāāāāāāāā
ā data/loader.py ā aligned daily features
āāāāāāāāāāā¬āāāāāāāāāāāā
ā¼
āāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāā
ā 5 independent signal modules ā each ā score in [-1, +1]
ā momentum(0.30) fear_greed(0.25) ā
ā funding(0.15) flows(0.15) vol(0.15) ā
āāāāāāāāāāāāāāāāāāāāāāāā¬āāāāāāāāāāāāāāāāāāāā
ā¼ weighted fusion
composite score
ā¼ priority rules
āāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāāā
ā 5-state regime classifier ā CAPITULATION > EUPHORIA >
ā (deterministic, no look-ahead) ā RISK_OFF > RISK_ON >
āāāāāāāāāāāāāāāāāāāāāāāā¬āāāāāāāāāāāāāāāāāāāā > RANGE_BOUND
ā¼
target exposure + action
āāāāāāāāāāāāāāāāā“āāāāāāāāāāāāāāāā
ā¼ ā¼
vectorized backtest MCP tool: get_market_regime
(no look-ahead, w/ costs) (stdio, Agent Hub skill)Signal | Source | Weight |
RSI / MACD momentum | CoinGecko price | 0.30 |
Fear & Greed Index | alternative.me | 0.25 |
Volatility regime | CoinGecko price | 0.15 |
Funding rate proxy | derived from price | 0.15 |
Exchange flow proxy | derived from volume | 0.15 |
Each signal outputs a normalized bullishness score in [-1, +1]. All 5 are independently unit-tested.
Transparency: Funding rate and exchange flows have no free public feed. We reconstruct them from price/volume data as clearly-labeled proxies. Drop in real feeds anytime ā the fusion layer is signal-agnostic.
git clone https://github.com/aggreyeric/bnb-market-regime-oracle.git
cd bnb-market-regime-oracle
pip install -r requirements.txt
# Run full pipeline: fetch ā classify ā backtest ā charts
python main.py
# Run tests (offline, no network needed)
PYTHONPATH=src python -m pytest tests/
# Run as MCP server
PYTHONPATH=src python -m market_regime_oracle.mcp_server
# Live MCP demo (30 seconds)
./scripts/demo.shdocker compose up --build run # full pipeline
docker compose up --build server # MCP servermarket_regime_oracle/
āāā src/market_regime_oracle/
ā āāā data/ # CoinGecko + alternative.me loaders
ā āāā signals/ # 5 signal modules (unit-tested)
ā āāā classifier/ # fusion ā regime mapping
ā āāā backtest/ # vectorized engine, no look-ahead
ā āāā viz/ # equity/drawdown/regime charts
ā āāā mcp_server.py # MCP stdio server
āāā tests/ # 24/24 passing
āāā results/ # CSVs, metrics.json, PNG charts
āāā scripts/demo.sh # live MCP round-trip
āāā Dockerfile
āāā docker-compose.yml
āāā README.mdCoinGecko v3 free API ā BTC daily close + volume
alternative.me ā Fear & Greed Index
Both public, both free. No API keys required.
MIT Ā© 2026
ā See CLAUDE.md for AI coding assistant context.
This server cannot be deployed
Crypto regime classifier for trading agents. 100% bounded regime accuracy. 50 free calls/day.
Crypto market intelligence: regime detection, funding rates, liquidations, prices, signals.
Live BTC/ETH risk state: risk policy, market structure and trading playbooks. Keyless free tier.
Stock market trading signals: regime, ranked buy list, stock + sector reads. Free tier + x402 live.
