smc-mcp
AkhileshSelvan/smc-mcp
Smart-money-concepts market analysis (BOS/CHoCH, order blocks, fair value gaps, liquidity sweeps) for any AI agent via the Model Context Protocol. Stocks, forex, gold, crypto — no API key.
Instalação
npx convez-ai AkhileshSelvan/smc-mcp1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
README
A tradução para o português deste README ainda está na fila.
SMC-MCP — Smart-Money-Concepts for AI agents
Give Claude, Cursor, or any MCP client the ability to read price action like a smart-money trader — market structure (BOS / CHoCH), order blocks, fair value gaps, and liquidity sweeps — for stocks, forex, gold, indices, and crypto.
No API key. No paid data feed. Ask in plain English:
"Run a full SMC read on XAUUSD on the 1h." "Are there any unmitigated bullish order blocks on EURUSD?" "Did BTCUSD sweep liquidity above the recent high?"
Every other finance MCP server fetches stock prices and standard indicators. This one speaks the language traders actually use: order blocks, FVGs, BOS, CHoCH, liquidity grabs.
30-second setup (Claude Desktop)
git clone https://github.com/AkhileshSelvan/smc-mcp.git
cd smc-mcp
pip install -e .
Add this to your claude_desktop_config.json
(Claude → Settings → Developer → Edit Config):
{
"mcpServers": {
"smc": {
"command": "python",
"args": ["-m", "smc_mcp"]
}
}
}
Restart Claude Desktop. You'll see the smc tools appear. Done.
Works the same in any MCP client (Cursor, Claude Code, etc.) — point it at
python -m smc_mcpover stdio.
Tools
| Tool | What it does |
|---|---|
smc_get_market_structure | Swing points, BOS / CHoCH events, current trend bias |
smc_find_order_blocks | Order blocks tied to structure breaks, with mitigation state |
smc_find_fair_value_gaps | Three-candle imbalances and whether they're filled |
smc_find_liquidity_sweeps | Swing highs/lows wicked through and rejected (stop hunts) |
smc_full_analysis | All of the above in one combined, plain-English read |
Every tool accepts:
symbol—AAPL,XAUUSD,EURUSD,BTCUSD,SPX… (trader shorthand is auto-translated)interval—5m,15m,30m,1h,1d,1wklimit— number of recent candles (20–1000)lookback— swing sensitivity (higher = only major swings)response_format—markdown(human read) orjson(machine-readable zones)
What the concepts mean
- BOS (Break of Structure) — price closes through a swing with the trend: a continuation signal.
- CHoCH (Change of Character) — price closes through a swing against the trend: the first hint of a reversal.
- Order block — the last opposite-colour candle before an impulsive move; an institutional footprint and a common entry zone.
- Fair value gap — a three-candle imbalance price often returns to rebalance.
- Liquidity sweep — a wick beyond an obvious swing that takes stops, then closes back (distinct from a BOS, which closes through).
Run the tests
pip install -e ".[dev]"
python tests/test_smc.py # or: python -m pytest tests/ -q
The core logic is pure Python over OHLC arrays (in src/smc_mcp/smc/), unit-
tested with hand-built series, and free of look-ahead bias — swings are only
used once their fractal has fully formed.
Roadmap
- Multi-timeframe confluence (HTF bias + LTF entry)
- Premium/discount (equilibrium) zones
-
ccxtdata source for exchange-native crypto candles - Optional MT5 data bridge for broker-accurate forex
- Prop-firm risk-gate tool (position size vs. drawdown rules)
Contributions and issues welcome.
Disclaimer
This software provides data analysis, not financial advice. Markets carry risk; validate every signal on your own charts and never trade on a tool's output alone.
License
MIT — see LICENSE.