The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Nyfed Markets listing page.
The New York Fed's Markets Data APIs — operation-level repo/reverse repo (ON RRP) auction results, securities lending operations, SOMA portfolio holdings, the Desk's reference rates (SOFR, EFFR, OBFR, TGCR, BGCR), and primary dealer survey statistics. This is the operation- and security-level data underneath FRED's daily rate aggregates, not a duplicate of them.
Part of Pipeworx — an MCP gateway connecting AI agents to 1683+ live data sources.
nyfed_repo_operations(operation_type, start_date, end_date, term, limit) — repo/reverse repo (ON RRP) auction results: totals, counterparty counts, award rates, by-security-type breakdown.nyfed_seclending_operations(operation, start_date, end_date, cusips, limit) — securities lending operation results (daily program + extended-term).nyfed_soma_holdings(view, holding_type, as_of_date, cusip, limit) — SOMA portfolio: summary by asset bucket over time, or full CUSIP-level Treasury/Agency snapshots by date, or one CUSIP's full holdings history.nyfed_reference_rates(rate_type, number) — SOFR/SOFRAI/TGCR/BGCR (secured) and EFFR/OBFR (unsecured), latest or recent history with percentile bands and volume.nyfed_primary_dealer_stats(mode, keyid_filter, seriesbreak, limit) — FR 2004 weekly primary dealer positioning/transaction survey; mode=list_series browses the keyid catalog, mode=latest returns current values.Keyless. No registration, no API key, no rate-limit headers observed.
https://markets.newyorkfed.org/api/rp/... — repo/reverse repo operations.https://markets.newyorkfed.org/api/seclending/... — securities lending operations.https://markets.newyorkfed.org/api/soma/... — SOMA holdings (summary, Treasury, Agency).https://markets.newyorkfed.org/api/rates/... — reference rates.https://markets.newyorkfed.org/api/pd/... — primary dealer statistics.https://markets.newyorkfed.org/static/docs/markets-api.yml (the .html doc page is a JS SPA that renders this file — fetch the .yml directly, the .html is useless to scrape)./api/rp/results/search.json's operationTypes query parameter is a confirmed SILENT NO-OP. Querying with operationTypes=Reverse Repo still returns Repo rows mixed in (verified live: 21 rows for a 9-day window that should have been ~7 RRP-only rows). method and term on that endpoint are unverified and not trusted either. This pack always filters operationType/term client-side after fetching. The path-based filter on the latest endpoint (/api/rp/{reverserepo|repo|all}/all/results/latest.json) DOES work correctly and is used for same-day queries instead./api/soma/asofdates/latest.json returned 2026-09-16 when probed on 2026-09-23 — normal, not a bug. as_of_date defaults to this latest published date.agency debts contains a literal space (/api/soma/agency/get/agency debts/asof/{date}.json). The tool accepts the friendly agency_debts and URL-encodes the space for you.holding_type=all on the treasury/agency SOMA endpoints returns EVERY CUSIP at that date — hundreds of rows. This pack truncates to limit (default 20) and always reports total_holdings so the truncation is visible rather than silent.PDPOSMBS-TOT) grouped into "series break" windows whose definitions can change over time (/api/pd/list/seriesbreaks.json is authoritative). This pack resolves "current" as the window whose enddate is 9999-12-31 rather than hardcoding a window id, so it keeps working across the next series break without a code change.effectiveDate are published the NEXT business day — EFFR/OBFR/TGCR/BGCR "today" have not been observed yet; only SOFR/SOFRAI sometimes show same-day. Not a lag bug.Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):
tools/list at https://gateway.pipeworx.io/nyfed-markets/mcp returns the tools in the table
above plus the shared Pipeworx meta-tools — ask_pipeworx,
discover_tools, search_within, remember/recall and the rest of the
gateway-wide set. So the tool count you see is larger than this table: a
single-pack endpoint currently lists roughly 30 shared tools alongside the
pack's own. The connection's initialize response states its exact scope, and
is the authoritative answer for a given day.
This is deliberate, not multiplexing by accident. The meta-tools are what let a
scoped connection answer a question this pack does not cover — via
ask_pipeworx, which routes across the whole catalog — without you adding a
second MCP server. There is currently no way to mount a pack endpoint without
them; if the extra schemas cost you more context than the routing is worth,
connect to the full gateway once rather than to several pack endpoints.
Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:
Both URLs reach the same gateway and the same 1683+ data sources. The
only difference is which pack's tools are listed directly; ask_pipeworx
reaches all of them from either one.
No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/nyfed_repo_operations. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.
This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:
Or run it directly to confirm it starts:
It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call
for only this pack's tools — none of the shared meta-tools the gateway
connection above adds. Same source, same tools, no ask_pipeworx routing.
Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:
The gateway picks the right tool and fills the arguments automatically.
MIT