US macro release calendar
Use this when you need official US release times (CPI, PPI, payrolls, GDP, PCE, retail sales, JOLTS, trade balance, productivity, claims, FOMC, ISM dates) with actual and prior values once released.
What it is
Official schedules from BLS, BEA, Census, the Federal Reserve, DOL (claims rule) and ISM (dates only). Actual/prior once released come only from the official publishers (keyless, no FRED): BLS Public Data API (CPI, PPI, payrolls, unemployment, JOLTS, ECI, productivity), BEA NIPA data files (GDP, PCE price), Census retail trade time series (retail sales), the DOL weekly claims release, the Federal Reserve Board (FOMC statement, target-range history) and the Census/BEA trade release on bea.gov.
Endpoints
- REST:
GET /v1/macro-calendaronhttps://api.cev-os.com - Legacy host (same service):
GET /v1/macro-calendaronhttps://cev-os.com - MCP tool:
get_macro_calendaronhttps://mcp.cev-os.com/mcp
Parameters
| Name | Type | Description | Constraints |
|---|---|---|---|
from | string | Start date (ET), YYYY-MM-DD. Default: today minus 30 days. | example 2026-10-01; pattern ^\d{4}-\d{2}-\d{2}$ |
to | string | End date (ET), YYYY-MM-DD. Default: today plus 60 days. | example 2026-11-10; pattern ^\d{4}-\d{2}-\d{2}$ |
event | string | Comma-separated event types: cpi, ppi, employment_situation, jolts, eci, gdp, pce, retail_sales, jobless_claims, fomc_decision, fomc_minutes, ism_manufacturing, ism_services, trade_balance, productivity | example cpi,fomc_decision |
status | string | released or scheduled | enum: released, scheduled |
limit | integer | Max events (1-1000, default 500). | min 1; max 1000 |
Example request
No account and no key. Query values below are the examples or defaults from the registry. An optional free key (header x-api-key) raises the limits; create one at https://cev-os.com/#api_keys after signing in with a verified email (Google sign-in is already verified).
curl -s "https://api.cev-os.com/v1/macro-calendar?from=2026-10-01&to=2026-11-10&event=cpi%2Cfomc_decision"
Example response
Every /v1 response is a JSON envelope (schema_version, product, timestamps, license, attribution, data). The registry example is the data member:
{
"count": 1,
"events": [
{
"id": "eci:2026-10-30",
"eventType": "eci",
"referencePeriod": "2026-Q3",
"releaseAt": "2026-10-30T12:30:00.000Z",
"releaseAtEt": "Fri, Oct 30, 2026, 8:30 AM ET",
"status": "released",
"actual": {
"value": 0.8,
"unit": "percent_3mo_sa",
"display": "+0.8% q/q"
},
"prior": {
"value": 0.9,
"unit": "percent_3mo_sa",
"display": "+0.9% q/q"
}
}
]
}
Update frequency
Cloud Scheduler: every 15 min 8:00-15:45 ET on weekdays and every 3 h otherwise; snapshot flagged stale after 6 h.
- Public cache:
300seconds (Cache-Control: public, max-age=300).
Sources, license and usage
License status: cleared. US Government works (public domain) plus ISM release dates. Cite the agencies; BEA non-endorsement notice and BLS API retrieval-date notice apply.
Attribution (show this wherever you display the data):
- Sources: U.S. Bureau of Labor Statistics (incl. BLS Public Data API), U.S. Bureau of Economic Analysis, U.S. Census Bureau, U.S. Department of Labor, Federal Reserve Board.
- This service uses BEA data but is not endorsed or certified by BEA.
- BLS.gov cannot vouch for the data or analyses derived from these data after the data have been retrieved from BLS.gov.
Privacy class: public-derived. Access: public. Price: free (price is null; pay-per-call is not active).
Gaps
- ISM uses its published 2026 table (live page is login-gated).
- Claims dates are rule-based.
- BLS, BEA and Census values are the current vintage (include revisions); claims actual is the advance figure as first published; trade prior is the revised previous month.
- Values can lag a release by up to 30 min (BLS API v1 limit 25 queries/day: one batched query per read, hard cap 20/day).
Related
- Data API hub (all products, limits, license notes)
- MCP Registry listing
com.cev-os/data - OpenAPI 3.1
- Machine product list:
/v1/products - Is it out yet? US macro calendar (explainer)
Reference data, not investment advice.