加密资产交易与AI代理平台

Crypto MCP Server 是一套本地化的 AI 对冲基金编排平台,将实时行情、技术与情绪分析、链上数据及多个 MCP 交易工具整合到一个仪表盘中,适用于需要自动化加密策略执行、回测和智能投研的加密资产用户。

  • by PelleNybe
  • TypeScript
  • 本地部署
  • 9 星标
  • 更新于
在 GitHub 查看 coraxgs/Crypto-MCP-Server---by-Corax-CoLAB

使用说明

🌌 Crypto MCP Server & Hedgehog Agentic Trading Dashboard 🦔

A Professional-Grade, Fully Localized AI Hedge Fund Orchestrator.

A massively scaled Model Context Protocol (MCP) infrastructure designed to act as a localized, AI-powered Hedge Fund Orchestrator. It acts as the backbone for local LLMs (like Claude Desktop) to ingest real-time market data, analyze technicals/sentiment, and autonomously execute trades.

🌟 Author: Pelle Nyberg | Corax CoLAB 🎪 Explore more: Crypto P's Crypto Circus

🧬 What makes this different?

This is not just another API wrapper. It is a multi-agent architectural playground.

  • Zero Mock-ups. 100% Real Data. Every visualizer, sonar sweep, and grid is driven by live websockets and REST data.
  • Extensible MCP Architecture. Easily add new modules. The system natively multiplexes JSON-RPC commands.
  • Hardened Security. Local SQLite, password-protected backend execution, no cross-site scripting gaps. Uses crypto.timingSafeEqual for password verification.

Our terminal is a living, breathing Hedge Fund AI Orchestrator, packed with cutting-edge tools.

🤖 Autonomous Orchestrator Mode (Agentic Loop)

Evolving from a passive Multi-MCP tool, the server acts as an autonomous 24/7 trading agent framework running a continuous Observe-Analyze-Act (OODA) loop.

  • 👁️ Observe (gather_market_data): Queries multiple MCPs (e.g., Technical indicators, News sentiment, On-chain data).
  • 🧠 Analyze (analyze_with_llm): Evaluates signals using an LLM (Gemini, Claude, or OpenAI) for a structured decision (BUY, SELL, HOLD).
  • ⚡ Act (execute_trade): Executes trades via the local CCXT MCP.
  • 🧪 Agentic Backtesting: Safely test AI prompts and models against historical OHLCV data without risking real funds.
  • 📖 Proof of Brain (Trading Diary): Comprehensive markdown reports detailing the "Board of Directors" reasoning for every action taken.

📱 Telegram Command Center

Receive real-time alerts and command your trading agent via Telegram.

  • /status: View active AI providers and last decisions.
  • /report: Receive your latest "Proof of Brain" report.
  • /analyze: Trigger a manual OODA cycle.

🌌 100% Real Data Integration & Visualizer Dynamics

The Crypto MCP Server uses real data across all visualizers. The entire system operates without a single mockup!

  • 🎯 Dark Pool Sonar: Real-time 3D sonar pings for large volume "whale" trades on central exchanges using @react-three/fiber and dynamic emissive materials.
  • 🔥 Flash-Crash Prediction Matrix: Glowing heatmap grid of bids to asks, tracking liquidity drains in real time via CCXT orderbooks.
  • 🚀 Galaxy View (Gravity Well): Cinematic 3D mapping of the top 50 cryptocurrencies with a glowing central sun and accretion disk.
  • 🧠 AI Sentiment Word-Cloud Sphere: Extracts trending keywords and sentiment from real-time news to form an interactive 3D floating sphere.
  • ⚡ Gas & Network Congestion Hologram: Visualizes Ethereum network congestion as a glowing, pulsating reactor core.
  • 🔬 Time-Machine Backtest Arena: Fully functional OHLCV visualizer to playback and simulate trading strategies (e.g. SMA Crossovers).
  • 🧠 Oracle Copilot: Voice-activated command center powered by local LLMs via MCP_LLM.
  • 🌍 Global Weather System: Dynamic 3D visualization of market volatility and network states using @react-three/fiber.
  • 🌊 Whale Sonar Sweep & Constellations: 3D plotting of significant market movers using CoinGecko and CCXT data.
  • 🌀 Arbitrage Wormhole: Scans and visualizes cross-exchange price discrepancies via MCP_CCXT.
  • 📊 Risk Radar Panel & Quantum Risk Map: Spatial representation of portfolio risk, exposure, and technical analysis indicators (RSI, MACD) via MCP_TA.
  • 📡 Market Sentiment Analyzer: Synthesizes news, price action, and LLM analysis for a comprehensive market mood score.
  • 🔮 Predictive Ghosting: Overlays projected price actions based on historical patterns using plotly and MCP_TA.
  • 🌐 Neural Net Liquidity & Trade Visualizer: 3D force-directed graphs showing the flow of liquidity and order execution.
  • 🏗️ Algo Grid Architect: Node-based strategy builder and visualizer for algorithmic trading logic.
  • 📈 Holo Order Flow & Topographic Order Book: Three-dimensional mapping of order book depth and flow.
  • 🪐 Asset Universe & Orbital Portfolio: Represents portfolio holdings as a 3D solar system with relative distances and sizes.
  • 📰 News Singularity: Aggregates and visually groups crypto news streams using MCP_NEWS.
  • ⚡ Volatility Matrix: Visualizes standard deviations and market swings in a grid.
  • 🖥️ System Overview: A comprehensive hub showing MCP node statuses and system health.

🚀 The Multi-MCP Ecosystem

This repository comes bundled with over a dozen powerful MCPs (Model Context Protocol). They provide execution, raw data, analysis, and external system integrations.

🛠️ View Included MCPs (Click to expand)
MCP Server Description Port
💱 ccxt_mcp The core exchange trading & market data integration via CCXT. 7001
⛓️ onchain_mcp ETH balances, ERC20 balances, transaction info, and live Gas Prices. 7002
📊 ta_mcp Compute Technical Indicators (RSI, MACD, SMA50, BB) and Monte Carlo Simulations. 7003
💼 portfolio_mcp Aggregated portfolio balances and performance tracking. 7004
🦎 coingecko_mcp Deep market stats, top coins, and historical trends. 7005
🧠 llm_mcp Local/Remote LLM operations and Copilot interactions. 7006
🔔 notifier_mcp System notifications, alerts, and Telegram broadcasting. 7007
📈 freqtrade_mcp Interfaces with local Freqtrade instances via REST APIs. 7011
🏟️ ha_mcp Prediction markets for agents via HeadlineArena (financial and Civic Index). 7018
🐙 octobot_mcp Interfaces with local OctoBot instances via REST APIs. 7012
🐦 hummingbot_mcp Controls local Hummingbot Gateway APIs. 7013
🤖 superalgos_mcp Interacts with the Superalgos platform API. 7014
📰 news_mcp Fetches the latest crypto news from CryptoPanic. 7017

🗺️ System Overview & Architecture

Explore the architecture, installation success, and security models below:

1️⃣ Architectural Overview (Click to expand)
Claude Desktop communicates via JSON-RPC with the Crypto MCP Server backend (REST + WebSocket). The server acts as a proxy, directing traffic to specific local MCP tools—such as CCXT, CoinGecko, and Portfolio—while logging orders to a local SQLite database.

📸 [INSERT LATEST GUI SCREENSHOT HERE] (Please upload the latest working system screenshot or animation here)

2️⃣ Installation and Configuration (Click to expand)
The terminal displays successful execution steps of the automated `install.sh` script, automating directory creation, Node.js installation, and service setup.
3️⃣ Security and Best Practices (Click to expand)
Summarizes the core security principles: using testnet keys, securing API keys, restricting network access, leveraging local control, and implementing an authenticated reverse proxy.

✅ Quick Start — Automated

Place the provided install.sh into $HOME/install.sh (or $HOME/cryptomcpserver/install.sh if you prefer). Make it executable and run it:

# Save install.sh to $HOME/install.sh, then:
cd $HOME
chmod +x install.sh
./install.sh

What install.sh does (summary):

  1. Creates directories and writes backend & frontend files.
  2. Installs Node.js if missing and runs npm install for backend & frontend.
  3. Ensures the orders table exists in $HOME/cryptomcpserver/gui/backend/orders.db.
  4. Frees port 4000 if occupied, then installs & enables the systemd service crypto-mcp-gui.service.
  5. Attempts a production build of the frontend.

After running, check service status and logs:

sudo systemctl status crypto-mcp-gui.service
sudo journalctl -u crypto-mcp-gui.service -f

🛠 Manual Install

If you prefer a hands-on approach:

1. Install system deps & Python Requirements:

sudo apt update
sudo apt install -y curl build-essential ca-certificates git python3-pip
pip3 install -r requirements.txt

2. Install Node.js (if needed):

curl -sL https://deb.nodesource.com/setup_20.x | sudo -E bash -
sudo apt install -y nodejs

3. Backend & Global Config:

cd $HOME/cryptomcpserver
cp .env.example .env
# Edit .env to add your passwords, keys, and allowed pairs

cd gui/backend
npm install

4. Frontend (Dev):

cd $HOME/cryptomcpserver/gui/frontend
npm install
npm run dev -- --host   # Open http://<PI_IP>:5173 on your laptop

5. Systemd (Backend):

# Create /etc/systemd/system/crypto-mcp-gui.service
sudo systemctl daemon-reload
sudo systemctl enable --now crypto-mcp-gui.service

⚙️ Configuration

The system uses a centralized .env file located at the root of the project to manage both Python MCP servers and the Node.js backend.

Copy and edit $HOME/cryptomcpserver/.env.example → .env:

# Essential configuration
PORT=4000
DASHBOARD_PASSWORD=your_secure_password # Required for trading and AI reasoning
ALLOWED_PAIRS=BTC/USDT,SOL/USDT # Fail-closed security: only these pairs are allowed
MAX_TRADE_USD=100.0 # Maximum allowed trade amount per transaction

# API Keys
BINANCE_API_KEY=your_key
BINANCE_API_SECRET=your_secret

# --- Autonomous Orchestrator Settings ---
ACTIVE_LLM_PROVIDER="gemini"
GEMINI_API_KEY="your_google_gemini_key_here"
TELEGRAM_BOT_TOKEN="your_token_here"
TELEGRAM_CHAT_ID="your_chat_id_here"

🔗 Claude Desktop Integration

Add MCP servers in Claude Desktop (Step-by-step)

  1. Open the Claude Desktop app.
  2. Navigate to App Settings / Preferences.
  3. Find Local MCP Servers.
  4. Click + (Add) and fill in the fields:
    • Name: ccxt
    • Description: CCXT MCP – exchange trading & market data
    • Transport: http
    • Endpoint: http://127.0.0.1:7001/mcp (if Claude runs on Pi) or http://<pi-ip>:7001/mcp (if Claude runs on laptop)
  5. Save. Repeat for other MCPs (coingecko, portfolio, onchain, ta, news, etc.) using their respective ports.

🖥 Dashboard User Manual

  • 📊 Portfolio: View aggregated balances & USD value.
  • 📈 Ticker: Live market data (via ccxt MCP).
  • 🛒 Order / Trade:
    • Preview (dry_run): Calculates estimated cost and logs a preview.
    • Confirm → Place order: Sends create_order to CCXT MCP.
  • 📜 Orders Log: Real-time updates via Socket.io, paginated with indices.
  • 🤖 AI Copilot: Voice-activated command center powered by local LLMs.

⚠️ Safety Warning: Always test with testnet keys. The UI requires confirmation to execute live orders.


🔒 Security & Best Practices

  • 🧪 Testnet First: Use testnet keys while testing your strategies.
  • 🔑 Environment Variables: Keep API keys absolutely out of the repository.
  • 🛡️ Network Isolation: Restrict access to MCP endpoints to your local LAN only.
  • 🔐 Authentication: Endpoints are fully secured with DASHBOARD_PASSWORD verification and robust anti-SSRF checks.

Stay Cypherpunk. Keep Building. ⚡