FRED Economic MCP Server

A powerful Model Context Protocol (MCP) server providing access to Federal Reserve Economic Data (FRED) through Claude and other MCP-compatible clients.

🌟 Features

• 📊 Comprehensive Data Access: Access to 800,000+ economic time series

• 🔍 Smart Search: Find economic indicators by keyword or category

• 📈 Historical Analysis: Retrieve up to 100,000 observations per series

• ⚡ High Performance: Efficient async operations with proper error handling

• 🎯 Economic Dashboard: Pre-configured key economic indicators

• 🔄 Multi-Series Support: Compare multiple economic series simultaneously

Related MCP server: FRED API MCP Server

🚀 Quick Start

Prerequisites

• Python 3.8 or higher

• FRED API key (free from FRED)

Installation

1. Download the server file:

   • Download fred_economic_server.py from this repository

2. Install dependencies:

pip install httpx python-mcp python-dotenv

3. Get your FRED API key:

   • Visit FRED API Key Registration

   • Sign up for a free account

   • Generate your API key

4. Set up environment:

export FRED_API_KEY="your_api_key_here"

Usage with Claude Desktop

Add to your Claude Desktop configuration (claude_desktop_config.json):

{
  "mcpServers": {
    "fred-economic": {
      "command": "python",
      "args": ["path/to/fred_economic_server.py"],
      "env": {
        "FRED_API_KEY": "your_api_key_here"
      }
    }
  }
}

🛠️ Available Tools

📊 Example Queries

Basic Data Retrieval

Ask Claude: "Get the latest GDP data"
Uses: get_fred_data("GDP")

Historical Analysis

Ask Claude: "Show me 10 years of unemployment data"
Uses: get_fred_historical("UNRATE", years=10)

Economic Research

Ask Claude: "Compare inflation and unemployment over the last 5 years"  
Uses: get_multiple_series(["CPIAUCSL", "UNRATE"])

📈 Common Economic Series

📞 Support

• 📧 Issues: Use GitHub Issues for bug reports and feature requests

• 📖 FRED API Docs: https://fred.stlouisfed.org/docs/api/

• 🔧 MCP Protocol: https://modelcontextprotocol.io/

📄 License

This project is licensed under the MIT License.