How to Connect FRED to Meta Muse via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → FRED, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

FRED + Meta Muse

Economic data from FRED (Federal Reserve Bank of St. Louis): search about 800,000 US and international series such as GDP, inflation, unemployment, interest rates and exchange rates, read values with transformations, browse releases and their calendar, and see revisions. Read-only, free API key.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

FRED_API_KEY

Step 1 — Get credentials

Setting up takes about 2 minutes and is free. You need a free FRED account.

1. Sign in or create a free account at https://fredaccount.stlouisfed.org.
2. Open API Keys (https://fredaccount.stlouisfed.org/apikeys), request a key and describe the use in one sentence. The key is a 32-character lower-case string and is issued at once.
3. Paste it into FRED_API_KEY. It is sent as the api_key query parameter.
4. Install the connector; the check reads the root category with fred_get_category. Then try fred_get_observations with UNRATE and limit 1 for the latest US unemployment rate.

Finding a series: fred_search_series searches titles, units, notes and tags; order_by popularity puts the headline series first, and search_type series_id matches ids with wildcards such as DGS*. Well-known ids: GDP, GDPC1 (real GDP), UNRATE, PAYEMS (payrolls), CPIAUCSL, CPILFESL (core CPI), PCEPI, FEDFUNDS, DGS10, T10Y2Y, MORTGAGE30US, SP500, VIXCLS, DCOILWTICO, DEXUSEU (USD per EUR), M2SL. You can also browse the category tree from the root category 0, list the series of a release, or combine tags such as gdp;germany with fred_list_tag_series.

What it reads: the values of a series, newest first (the latest 1000 by default, or a period with observation_start and observation_end); the metadata of a series (units, frequency, seasonal adjustment, notes, last update) with its categories, release and tags; categories and their series; statistical releases, their series and publication dates, and the calendar across all releases, including upcoming dates; data sources and their releases; tags; recently updated series; and the vintage dates on which a series was published or revised. The data comes from more than 100 sources such as BEA, BLS, the Federal Reserve Board, the Census Bureau, OECD, IMF and Eurostat.

Transformations: units turns levels into change, percent change, percent change from a year ago (pc1), compounded annual rates or a natural log. frequency with aggregation_method (average, sum or end of period) aggregates to a lower frequency, e.g. daily rates to monthly averages; it cannot be higher than the series' own frequency.

Revisions: fred_list_vintage_dates lists when a series was published or revised (ALFRED). fred_get_observations with realtime_start and realtime_end set to the same vintage date shows the data as it was known on that day, e.g. the first estimate of a GDP quarter.

Data formats: values are strings in the units of the series, so the AI checks units and seasonal adjustment before comparing two series. A value of "." means no observation for that date, e.g. a holiday. Dates are YYYY-MM-DD and mark the start of the period (2026-04-01 for the second quarter of 2026). Monthly and quarterly series come with a lag, so the AI quotes the date of the latest observation, not today. The window of fred_list_series_updates is YYYYMMDDHhmm in US Central time. The AI cites FRED and the original source from the series notes; some series are copyrighted by their source, and the notes say so.

Read-only: every tool only reads, so there is nothing to confirm and no tool is switched off at install.

Limits: up to 120 requests per minute per key. Lists page with limit and offset.

Common errors: FRED answers 400 with an error_message that names the problem: "api_key is not registered" for a wrong or missing key, "The series does not exist" for an unknown id, or an invalid parameter such as a frequency higher than the series' own. 429 means the rate limit was hit: wait a minute.

The connector does not cover GeoFRED maps and regional shapes, the bulk FRED v2 release download or series groups.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Muse

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Muse adds remote MCP servers as custom connectors and signs in to them with OAuth.

1. In Muse, open Settings → Connectors and choose Add custom connector.
2. Paste the MCP server URL https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID.
3. Muse opens the AnythingMCP sign-in page: sign in and approve access.
4. In a chat, ask Muse for something that needs the FRED tools.

You can also set it up from a chat, on the web, in the iOS and Android apps or in WhatsApp. Send Muse this message:

Create a Custom Connector for a new remote MCP server, then connect to it:
Name: AnythingMCP
Transport: remote streamable HTTP
URL: https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID
Auth: OAuth

Self-hosting? Use your instance's public HTTPS address instead of https://cloud.anythingmcp.com. Muse connects from Meta's cloud, so localhost does not work.

Available tools

Example prompts

• "What is the latest US unemployment rate in FRED, and which month does it refer to?"
• "Show US CPI inflation year over year for the last 24 months."
• "Compare the 10-year and 2-year Treasury yields over the past year as monthly averages."
• "When is the next Employment Situation release scheduled?"
• "What was the first published value of US real GDP for the first quarter of 2026, and how has it been revised since?"

FAQ

Where is Muse available? Meta currently offers Muse in the US and Canada, on the web, in the iOS and Android apps and in WhatsApp.

Does Meta review this connector? No. Meta doesn't review custom connectors. Muse sees only the tools you switched on for this MCP server in AnythingMCP, and you can disconnect it anytime under Settings → Connectors.

Next steps

• Connector landing page
• Connect to Claude
• Connect to ChatGPT
• Connect to Cursor
• Connect to Gemini
• Connect to Microsoft Copilot