@token-metrics-ai/mcp

Token Metrics MCP Server

The Token Metrics Model Context Protocol (MCP) server provides comprehensive cryptocurrency data, analytics, and insights through function calling. This server enables AI assistants and agents to access Token Metrics' powerful API for real-time crypto market data, trading signals, price predictions, and advanced analytics.

Features

• Real-time Crypto Data: Access current prices, market cap, volume, and other key metrics
• Trading Signals: AI-generated trading signals for long and short positions
• Price Predictions: Advanced price forecasting and scenario analysis
• Technical Analysis: Support and resistance levels, correlation analysis
• Market Analytics: Comprehensive market insights and sentiment analysis
• Quantitative Metrics: Advanced quantitative analysis and grading systems

Quick Start

Using npx (Recommended)

The easiest way to get started is using npx:

# Set environment variable and run
export TOKEN_METRICS_API_KEY=your_api_key_here
npx -y @token-metrics-ai/mcp@latest

Setup with AI Clients

Claude Desktop or VS Code/Cursor

Add the following to your claude_desktop_config.json or mcp.json:

{
  "mcpServers": {
    "token-metrics": {
      "command": "npx",
      "args": ["-y", "@token-metrics-ai/mcp@latest"],
      "env": {
        "TOKEN_METRICS_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Available Tools

The Token Metrics MCP server provides the following tools:

Token Data & Prices

• get_tokens_data - Fetch comprehensive token information
• get_tokens_price - Get current token prices
• get_tokens_hourly_ohlcv - Hourly OHLCV data
• get_tokens_daily_ohlcv - Daily OHLCV data

Trading & Analysis

• get_tokens_trading_signal - AI-generated trading signals
• get_tokens_trader_grade - Short-term trader grades
• get_tokens_investor_grade - Long-term investor grades
• get_tokens_resistance_and_support - Technical support/resistance levels
• get_tokens_correlation - Token correlation analysis

Market Intelligence

• get_market_metrics - Overall market analytics
• get_sentiment - Market sentiment analysis
• get_tokens_quant_metrics - Quantitative metrics
• get_tokens_scenario_analysis - Price prediction scenarios

Research & Reports

• get_tokens_ai_report - AI-generated token reports
• get_crypto_investors - Crypto investor information
• get_top_tokens_by_market_cap - Top tokens by market cap

Indices & Portfolio

• get_indices - Fetch active and passive crypto indices
• get_indices_performance - Historical performance data for indices
• get_indices_holdings - Current holdings and weights for indices

Getting Your API Key

1. Visit Token Metrics
2. Sign up for an account
3. Navigate to your API Dashboard
4. Generate a new API key
5. Use the API key with this MCP server

Development

Prerequisites

• Node.js 18 or higher
• npm or yarn
• TypeScript

Local Development

1. Clone the repository:

git clone https://github.com/token-metrics/mcp.git
cd mcp

2. Install dependencies:

npm install

3. Set your API key:

export TOKEN_METRICS_API_KEY=your_api_key_here

4. Run in development mode:

npm run start:dev

Building

npm run build

Testing with MCP Inspector

You can test the server using the MCP Inspector:

# Build the server first
npm run build

# Run with MCP Inspector
npx @modelcontextprotocol/inspector node build/src/cli.js

Configuration

The server accepts the following configuration options:

• --help - Show help information

Environment variables:

• TOKEN_METRICS_API_KEY - Your Token Metrics API key

Error Handling

The server includes comprehensive error handling:

• Invalid API Key: Returns authentication error
• Rate Limiting: Handles API rate limits gracefully
• Network Issues: Retries failed requests
• Invalid Parameters: Validates input parameters

Security

• API keys are handled securely
• No sensitive data is logged
• Docker container runs as non-root user
• Input validation on all parameters

Support

• Documentation: Token Metrics API Docs
• Issues: GitHub Issues
• Support: Token Metrics Support

License

MIT License - see LICENSE file for details.