Federal Register Scraper — Rules, Notices & Agency Documents

Scrape the Federal Register — every final rule, proposed rule, notice and presidential document published by US federal agencies — straight from the official Federal Register API. No API key, no login, no headless browser. Run it empty and you get the newest documents; add filters and you get exactly the slice of the regulatory record you care about.

Coverage runs back to 1994, across 472 agencies, with every document addressed by its permanent Federal Register document number.

Quick Start: Run in 1 Line

Python (using reapx SDK)

import os, reapx
os.environ["APIFY_TOKEN"] = "usr_your_apify_token" # Free $5 monthly credit included
results = reapx.live("federal-register-scraper", {"searchTerm": "artificial intelligence"})

cURL

curl -X POST "https://api.apify.com/v2/acts/reapx~federal-register-scraper/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
     -H "Content-Type: application/json" \
     -d '{"searchTerm": "artificial intelligence", "maxDocuments": 20}'

What this Federal Register scraper returns

One row per document:

Field	What it gives you
documentNumber, title, htmlUrl, pdfUrl	The document and permanent links to it
documentTypeLabel, subtype	Final rule, proposed rule, notice or presidential document
agencyNames, agencySlugs, topics	Who issued it and what it is about
publicationDate, effectiveOn, signingDate	When it published, takes effect, was signed
commentsCloseOn, commentsOpen, daysUntilCommentsClose, commentUrl	The comment window and where to file
significant	Flagged under Executive Order 12866 as economically material
docketIds, regulationIdNumbers, cfrReferences	Docket, RIN and the CFR parts it touches
citation, volume, startPage, endPage, pageLength	Formal citation and how long it runs
abstract, action, fullText	The summary, the action line, and optionally the whole body

Five named dataset views ship with it: Documents, Open rulemaking, Effective dates, Presidential documents and Citations & pages.

How to scrape the Federal Register

1. Run it with no input at all — you get the 100 newest documents.
2. Narrow by agency (environmental-protection-agency), document type, or a search term.
3. Add a date window, a CFR part, a docket ID or a RIN to target one rulemaking.
4. Set Max documents. You are charged once per document returned, so that is your cost cap.

Filters

• Search term — full text across title, abstract and body
• Document types — final rule, proposed rule, notice, presidential document
• Agencies — any of the 472 agency slugs, several at once
• Subject sections — money, business and industry, environment, health and public welfare, science and technology, world
• Published / effective date windows, or a rolling "last N days"
• Open for public comment only and comments close before — the actionable docket
• Significant regulatory actions only — EO 12866
• CFR title and part, docket ID, regulation ID number (RIN)
• Presidential document types and presidents — executive orders, proclamations, memoranda

What you can build with it

• Regulatory monitoring — run daily filtered to your agencies and diff by documentNumber.
• Comment-deadline tracking — openForCommentOnly plus daysUntilCommentsClose is a live docket calendar.
• Compliance calendars — effectiveFrom gives you every rule coming into force in a window.
• Rulemaking histories — one RIN returns every document in that rulemaking, in order.
• Executive order archives — all presidential documents by president or by type.
• Policy datasets for LLMs and RAG — includeFullText gives clean plain text with a citation.

Pricing

Pay per event: $0.002 per document returned, with automatic discounts on every Apify plan tier (Bronze through Diamond). Nothing else is billed to you — no separate platform-usage line.

You are charged only for complete document rows written to the dataset. A filter combination the API rejects, an unreachable API, and a query that matches nothing all cost zero. Set Max total charge on the run to cap spend; the actor checks that cap before it starts and stops cleanly the moment it is reached.

Input example

{
  "documentTypes": ["PRORULE"],
  "agencies": ["environmental-protection-agency"],
  "openForCommentOnly": true,
  "maxDocuments": 200
}

Output example

{
  "documentNumber": "2026-15372",
  "title": "Air Plan Approval; Pennsylvania; Redesignation of the Warren Area",
  "documentType": "Proposed Rule",
  "documentTypeLabel": "Proposed rule",
  "abstract": "EPA is approving a state implementation plan revision submitted by Pennsylvania.",
  "action": "Proposed rule.",
  "agencyNames": ["Environmental Protection Agency"],
  "agencySlugs": ["environmental-protection-agency"],
  "topics": ["Air pollution control", "Ozone"],
  "publicationDate": "2026-08-03",
  "effectiveOn": null,
  "commentsCloseOn": "2026-08-31",
  "commentsOpen": true,
  "daysUntilCommentsClose": 28,
  "significant": false,
  "citation": "91 FR 37155",
  "docketIds": ["EPA-R03-OAR-2025-0321"],
  "regulationIdNumbers": [],
  "cfrReferences": ["40 CFR 52"],
  "htmlUrl": "https://www.federalregister.gov/documents/2026/08/03/2026-15372/air-plan-approval-pennsylvania",
  "daysSincePublication": 0,
  "scrapedAt": "2026-08-03T09:15:00Z"
}

FAQ

How do I scrape the Federal Register?

Run this actor with no input and it returns the newest documents. Everything comes from the official Federal Register API at federalregister.gov/api/v1, which is public and needs no key.

Do I need an API key or an account?

No. The Federal Register API is open. There is no authentication step and no private data involved.

How do I find all documents from one agency?

Put the agency slug into Agencies — environmental-protection-agency, food-and-drug-administration, securities-and-exchange-commission. The full list of 472 slugs is at federalregister.gov/api/v1/agencies.json, and every row this actor returns carries agencySlugs you can feed straight back in.

How do I track public comment deadlines?

Turn on Open for public comment only. Every row then carries commentsCloseOn, daysUntilCommentsClose and, where the agency published one, a commentUrl for filing.

How far back does the data go?

To 1994. Set Published from to any date in that range, or sort oldest first to walk forward from the beginning.

How do I follow a single rulemaking from proposal to final rule?

Use the Regulation ID number (RIN). One RIN returns every document in that rulemaking across its whole life. A Docket ID does the same for the agency's own docket numbering.

What is a "significant" regulatory action?

A document flagged under Executive Order 12866, broadly the economically material ones. Turn on Significant regulatory actions only to keep just those.

Can I get the full text of each document, not just the abstract?

Yes — turn on Include the full document text. It adds a fullText column with the complete plain-text body up to 40,000 characters. It costs one extra request per document, so large runs take longer, but the price per document does not change.

How many documents can one run return?

Up to 10,000 per query, which is the API's own ceiling for a single filter set. Narrow the date window and run again to walk past it.

Can an AI agent call this actor?

Yes. Every input and output field carries a title, description and example, and every row is addressed by its documentNumber, so an agent can fill the input and chain the output without extra prompting.

---

Unofficial - not affiliated with the Federal Register. Collects public data only. reapx. Contact reapxdev@proton.me.