@Blockrun/Alpha

AI-powered crypto trading MCP server with technical analysis, sentiment, and execution tools.

Installation

claude mcp add alpha npx @blockrun/alpha

Tools

Usage Examples

> "Get ETH technical signals"
Uses alpha_signal to fetch RSI, MACD, EMA indicators

> "Check PEPE market data"
Uses alpha_dex for DexScreener data

> "Find similar trades to ETH breakout"
Uses alpha_memory for semantic search

> "Check if $500 ETH buy passes risk limits"
Uses alpha_risk to validate against position limits

> "Swap 100 USDC for ETH"
Uses alpha_swap with execute=true to swap on Base

Data Sources

• CoinGecko API - Price data and technical indicators (free)
• DexScreener - DEX market data (free)
• Swap Aggregators - Uses first available:
  • 1inch (set ONEINCH_API_KEY)
  • 0x (set ZEROX_API_KEY)
  • Paraswap (no key required, default fallback)
• BlockRun Twitter - Sentiment analysis (paid, requires wallet)

Environment Variables

Risk Limits (Hardcoded)

Data Storage

Data is stored locally in ~/.blockrun/alpha/alpha.db:

• Portfolio positions
• Trade history
• Trade memory with vector embeddings

Requirements

• Node.js 18+
• BlockRun wallet (optional, for sentiment and swaps)

Extend with More MCPs

Use alpha_extensions to discover additional capabilities:

# Example: Add Nansen for on-chain data
claude mcp add --transport http nansen https://mcp.nansen.ai/ra/mcp \
  --header "NANSEN-API-KEY: YOUR_KEY"

Related

• @blockrun/mcp - AI model access
• alpha-trader - Claude Code trading agent

License

MIT