FRED MCP server for AI agents
Let Claude, ChatGPT or your own agent read FRED: 6 typed functions through one MCP server. No FRED API key needed.
- Functions
- 6
- Sign-in
- Not needed
- Works with
- MCP, HTTP, Node, Python
Connect
An MCP server for FRED alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.
https://api.bowmark.ai/mcp/site/fredOne MCP server for FRED and every other site Bowmark covers. Your agent looks up what it needs and calls it.
https://api.bowmark.ai/mcpSet it up in your clientBoth need a Bowmark API key. The quickstart gets you one and a first result in a few minutes.
Functions
Open a function to see its arguments, what it returns and how to call it.
browseCategoryBrowses FRED's category tree the way fred.stlouisfed.org/categories does.Arguments
Name Type Required categoryIdnumberNo Returns
fredCategoryinterface fredSeries { id: string; title: string; units: string; unitsShort: string; frequency: string; frequencyShort: string; seasonalAdjustment: string; seasonalAdjustmentShort: string; observationStart: string; observationEnd: string; lastUpdated: string; popularity: number | null; notes: string | null; } interface fredCategoryRow { id: number; name: string; parentId: number; } interface fredCategory { id: number; name: string; parentId: number; children: fredCategoryRow[]; series: fredSeries[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/fred/browseCategoryCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.fred.browseCategory();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.fred.browseCategory())curl -X POST https://api.bowmark.ai/v1/sites/fred/browseCategory \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.fred.browseCategory();getSeriesInfoReads the metadata for one known series id.Arguments
Name Type Required seriesIdstringYes Returns
fredSeriesinterface fredSeries { id: string; title: string; units: string; unitsShort: string; frequency: string; frequencyShort: string; seasonalAdjustment: string; seasonalAdjustmentShort: string; observationStart: string; observationEnd: string; lastUpdated: string; popularity: number | null; notes: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/fred/getSeriesInfoCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.fred.getSeriesInfo("GDPC1");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.fred.getSeriesInfo("GDPC1"))curl -X POST https://api.bowmark.ai/v1/sites/fred/getSeriesInfo \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"seriesId":"GDPC1"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.fred.getSeriesInfo("GDPC1");getSeriesObservationsReads the actual numbers for a FRED series.Arguments
Name Type Required argsstring | { seriesId: string; from?: string; to?: string; units?: string; frequency?: string; aggregationMethod?: string }Yes Returns
fredObservationsinterface fredObservation { date: string; /** null where FRED holds no observation for that period, a market holiday in a * daily series, or the first year of a year-over-year transform. */ value: number | null; } interface fredObservations { seriesId: string; /** The transform FRED reports having applied: "lin" (none), "pc1", "pch", … */ units: string; observationStart: string | null; observationEnd: string | null; /** null: the public CSV has no vintage envelope. */ realtimeStart: string | null; realtimeEnd: string | null; count: number; observations: fredObservation[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/fred/getSeriesObservationsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.fred.getSeriesObservations("GDPC1");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.fred.getSeriesObservations("GDPC1"))curl -X POST https://api.bowmark.ai/v1/sites/fred/getSeriesObservations \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":"GDPC1"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.fred.getSeriesObservations("GDPC1");listReleasesLists FRED's economic data releases.Arguments
Name Type Required args{ releaseId?: number; from?: string; to?: string }No Returns
fredRelease[]interface fredReleaseDate { date: string; time: string | null; released: boolean; } interface fredRelease { id: number; name: string; url: string; dates: fredReleaseDate[] | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/fred/listReleasesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.fred.listReleases();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.fred.listReleases())curl -X POST https://api.bowmark.ai/v1/sites/fred/listReleases \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.fred.listReleases();listSourcesLists the agencies and organizations that originate FRED's data.Arguments
Name Type Required args{ sourceId?: number }No Returns
fredSource[]interface fredSource { id: number; name: string; url: string; releases: { id: number; name: string; url: string }[] | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/fred/listSourcesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.fred.listSources();import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.fred.listSources())curl -X POST https://api.bowmark.ai/v1/sites/fred/listSources \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.fred.listSources();searchSeriesFinds the economic data series that match a search text.Arguments
Name Type Required argsstring | { query: string; limit?: number }Yes Returns
fredSeries[]interface fredSeries { id: string; title: string; units: string; unitsShort: string; frequency: string; frequencyShort: string; seasonalAdjustment: string; seasonalAdjustmentShort: string; observationStart: string; observationEnd: string; lastUpdated: string; popularity: number | null; notes: string | null; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/fred/searchSeriesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.fred.searchSeries("gdp");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.fred.searchSeries("gdp"))curl -X POST https://api.bowmark.ai/v1/sites/fred/searchSeries \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":"gdp"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.fred.searchSeries("gdp");
Use it from code
- Node
npm i @bowmark/web - Python
pip install bowmark-web - HTTP
POST /v1/sites/fred/<function> - OpenAPI
/v1/sites/fred/openapi.json