Tools / Federal Register Scraper – Agency, Type & Comment Deadlines

Federal Register Scraper – Agency, Type & Comment Deadlines

Scrape the US Federal Register's official government API: final rules, proposed rules, notices and presidential documents. 30 flat fields incl. comment-close deadline, EO 12866 significance, RIN, docket IDs and CFR references, with cursor paging past the API's own 10,000-row wall.

Run it

curl -X POST https://fetchsmith.com/api/v1/run/federal-register-scraper \
  -H "Authorization: Bearer YOUR_KEY" -H "Content-Type: application/json" \
  -d '{"agencies": ["Environmental Protection Agency"], "commentsOpenOnly": true, "documentTypes": ["RULE", "PRORULE"], "maxResults": 100}'

Output fields

documentNumber type subtype title abstract action publicationDate effectiveOn commentsCloseOn signingDate datesText significant agencyNames agencySlugs parentAgencyNames docketIds regulationIdNumbers cfrReferences topics citation startPage endPage pageLength president executiveOrderNumber excerpt url pdfUrl fullTextUrl jsonUrl

Sample output

{
  "agencyNames": [
    "Environmental Protection Agency"
  ],
  "cfrReferences": [
    "40 CFR 257"
  ],
  "citation": "91 FR 57842",
  "commentsCloseOn": "2026-11-10",
  "docketIds": [
    "EPA-HQ-OLEM-2026-4324",
    "FRL-13374-01-OLEM"
  ],
  "documentNumber": "2026-18552",
  "fullTextUrl": "https://www.federalregister.gov/documents/full_text/text/2026/09/11/2026-18552.txt",
  "publicationDate": "2026-09-11",
  "title": "Wisconsin: Approval of State Coal Combustion Residuals Permit Program",
  "type": "Proposed Rule",
  "url": "https://www.federalregister.gov/documents/2026/09/11/2026-18552/wisconsin-approval-of-state-coal-combustion-residuals-permit-program"
}

Guides

The Federal Register API says it has 10,000 documents. It doesn't — and the fix is already in the response — Every US federal rule, proposed rule and notice since 1994 is free JSON with no API key. But `count` is clamped, offset paging 400s at row 10,000, and a mistyped agency slug kills the whole query instead of returning nothing.

Only publicly accessible pages are read. Respect the source site's terms when using the data.