Yes. The server is open source under the AGPL-3.0 license, and FRED data is free. You need a free FRED API key from the St. Louis Fed to authenticate. Commercial licensing is available from the author for closed-source use.

Does it support remote/OAuth?

It runs locally over stdio via npx and can also run over a streamable-HTTP transport for self-hosting. It authenticates with a FRED API key rather than OAuth, since FRED uses key-based access.

How many series does it cover?

FRED hosts over 800,000 economic time series, all reachable through this server's three tools. fred_browse and fred_search help locate the right series before fred_get_series retrieves its observations.

FRED MCP server

CommunityStefano Amorelli98Config last verified Jun 1, 2026

A maintained MCP server for FRED that gives agents access to all 800,000+ Federal Reserve economic data series through three focused tools.

FRED MCP is an actively maintained community server that connects an AI agent to FRED (Federal Reserve Economic Data), the St. Louis Fed's free database of more than 800,000 economic time series. With it a model can answer macro questions with real numbers — GDP, unemployment, inflation, interest rates, money supply, exchange rates, and far more — instead of relying on stale memorized figures. It is built around a small, deliberate tool surface rather than dozens of endpoint wrappers, which keeps the agent's choices clear: browse the catalog, search for the right series, then retrieve its data.

The three tools are fred_browse to explore FRED's catalog by categories, releases, sources, and series; fred_search to find series by keywords, tags, and filters; and fred_get_series to retrieve a series' observations with date-range and transformation options (such as percent change or year-over-year). The server is published as the fred-mcp-server npm package and runs locally over stdio via npx, authenticating with a free FRED_API_KEY from the St. Louis Fed. It can also run over a streamable-HTTP transport if you want to host it. It is AGPL-3.0 licensed with commercial licensing available.

Quick install

Copy-paste configs are provided for all 8 supported clients. Pick your client below.

Add to ~/.claude.json

~/.claude.json

json

{
  "mcpServers": {
    "fred": {
      "command": "npx",
      "args": [
        "-y",
        "fred-mcp-server"
      ],
      "env": {
        "FRED_API_KEY": "<FRED_API_KEY>"
      }
    }
  }
}

Or via CLI

bash

claude mcp add fred -- npx -y fred-mcp-server

Step-by-step guides

Available tools

Tool	Description
fred_browse	Explores FRED's catalog by categories, releases, sources, and series.
fred_search	Finds economic data series by keywords, tags, and filtering criteria.
fred_get_series	Retrieves a series' observations with date-range and transformation options.

Required configuration

FRED_API_KEYRequired
Free FRED API key from the St. Louis Fed, used to authenticate requests.

What you can do with it

Answer macro questions with live data

The agent searches for the right series, then pulls current CPI, unemployment, or GDP observations to summarize the macro backdrop without a human assembling the figures by hand.

Build a transformed time series

Use fred_get_series with a transformation like year-over-year percent change over a date range to feed an analysis or chart directly from the source data.

FAQ

Is it free?: Yes. The server is open source under the AGPL-3.0 license, and FRED data is free. You need a free FRED API key from the St. Louis Fed to authenticate. Commercial licensing is available from the author for closed-source use.
Does it support remote/OAuth?: It runs locally over stdio via npx and can also run over a streamable-HTTP transport for self-hosting. It authenticates with a FRED API key rather than OAuth, since FRED uses key-based access.
How many series does it cover?: FRED hosts over 800,000 economic time series, all reachable through this server's three tools. fred_browse and fred_search help locate the right series before fred_get_series retrieves its observations.

← Browse all finance servers