AnswerPool

EDGAR Text Search API: Search EDGAR filing text for any phrase with form and date filters — SEC EDGAR

Search EDGAR filing text for any phrase with form and date filters — returns matching filings with filer, date, items and URL. The search competitors gate at their top tier or lack entirely; one call, no crawling.

$0.02 per call — pay with prepaid card credits (Authorization: Bearer ck_live_…, packs from $1) or with USDC over x402: call it, get 402 with the price, retry signed. Failed calls are never charged.

How do you call it?

GET /v1/sec/search/fulltext — product id sec_full_text_search. Over MCP, call answerpool_get with product_id=sec_full_text_search.

curl "https://answerpool.io/v1/sec/search/fulltext?q=supply+chain+disruption&forms=8-K&start=2026-08-01&end=2026-08-31"

# with a prepaid credit key
curl -H "Authorization: Bearer ck_live_..." "https://answerpool.io/v1/sec/search/fulltext?q=supply+chain+disruption&forms=8-K&start=2026-08-01&end=2026-08-31"

Live call: https://answerpool.io/v1/sec/search/fulltext?q=supply+chain+disruption&forms=8-K&start=2026-08-01&end=2026-08-31 · JSON sample: /v1/samples/sec_full_text_search

What does the answer look like?

A representative response, the same static sample served at /v1/samples/sec_full_text_search.

as_of2026-08-31T14:00:00Z
count1
refresh_after2026-09-17T11:15:00.000Z

filings

accessioncikcompanyformfileditemsurl
0001437749-26-0289200000320193APPLE INC8-K2026-08-291.03https://www.sec.gov/Archives/edgar/data/320193/000143774926028920-index.htm
The same sample as raw JSON
{
 "as_of": "2026-08-31T14:00:00Z",
 "count": 1,
 "filings": [
  {
   "accession": "0001437749-26-028920",
   "cik": "0000320193",
   "company": "APPLE INC",
   "form": "8-K",
   "filed": "2026-08-29",
   "items": [
    "1.03"
   ],
   "url": "https://www.sec.gov/Archives/edgar/data/320193/000143774926028920-index.htm"
  }
 ],
 "refresh_after": "2026-09-17T11:15:00.000Z"
}

How fresh is it, and where does the data come from?

The fastest source behind this answer can change every 15 minutes, so every response carries a refresh_after timestamp — right now it would be 2026-09-17T11:15:00.000Z. Schedule the next call on that value rather than on a guess.

Method version 0.1.0 · serving cache 1 day · every response carries a result_id you can resolve at /v1/provenance.

When should you use it?

Content questions across filings.

What you skip building

When should you not use it?

Metadata-only listing — the metadata search is cheaper.

Where else can you find this?