REST & CSV

The Scrolls Data API returns tabular JSON for notebooks and scripts, plus CSV mirrors for Excel From Web.

Base URL

http://localhost:8000

JSON response shape

{
  "columns": ["date", "GDP"],
  "rows": [["2020-01-01", 21427.7], ["2020-04-01", null]]
}

Catalog

curl -s http://localhost:8000/fred/catalog

Series (wide table)

curl -s -X POST http://localhost:8000/fred/series \
  -H 'Content-Type: application/json' \
  -d '{"series_ids": ["GDP", "UNRATE"], "start_date": "2020-01-01"}'

CSV

Endpoint	Example
`GET /fred/catalog.csv`	Full catalog
`GET /fred/series.csv`	`?series_ids=GDP,UNRATE&start_date=2020-01-01`

Cross-source search

Unified semantic search fans out across macro sources and EDGAR 8-K. Use POST /search (SSE). The Scrolls site proxies it at /api/data/search. For institutional clients, prefer the Python SDK over gRPC.

curl -N -X POST http://localhost:8000/search \
  -H 'Content-Type: application/json' \
  -H 'Accept: text/event-stream' \
  -d '{"q":"policy rate cuts","sources":["fred","edgar_8k"],"limit":10}'

Stream events: source_result / source_error / done. GraphQL query/subscription twin: POST /graphql.

Sources

Routes follow the same shape under /edgar, /fred, /bls, /census, /eurostat, /imf, and /bis.