US Copyright Office Scraper

Scrape the US Copyright Office Public Records System (COINS) — the official public catalog of copyright registrations, recordations and card-catalog records (1938–today, 60M+ records). Search by keyword / title, registration number, owner/claimant or author, and get the full registration record: registration numbers, titles, claimants, authors, dates of creation/registration/publication, type of work, renewal info, correspondence addresses, and card-catalog images. HTTP-only via the public COINS JSON API. No auth, no proxy required.

What this actor does

• Four modes: search, searchByNumber, searchByOwner, searchByAuthor
• Full record detail: every emitted record is enriched with the official detailed-record payload (claimants with addresses, author statements, imprint, basis of claim, renewal/previous-registration info, dates)
• Filters: type of work, registration class (TX/PA/SR/VA/RE/MW/PRE...), registration status, record type (registration/recordation), recordation item type, system of origin, date range on 6 date axes, must-contain keyword, sort
• Card-catalog coverage: pre-1978 card-catalog records include scanned card images from the Library of Congress tile server
• Empty fields are omitted

Output per record

• publicRecordsId — COINS record ID (e.g. voyager_16089852, card_catalog_...)
• title, titleProper, variantTitles[]
• registrationNumber, copyrightNumber, registrationNumbers[], cancelledRegistrationNumbers[], previousRegistrationNumbers[]
• registrationClass, registrationStatus, typeOfWork, typeOfWorkEnglish, allTypesOfWork[]
• claimants[], claimantAddresses[], primaryClaimant, authors[], authorDetails[] (domicile, citizenship, authorship), primaryAuthor
• registrationDate, dateOfCreation, dateOfPublication, publicationYear, representativeDate, applicationDate, depositReceivedDate
• basisOfClaim[], imprint[], physicalDescription, preexistingMaterial, materialExcluded, rightsAndPermissions, copyrightNote
• Recordations: recordationNumber, recordationDate, recordationItemType, recordationNumberRange, numberOfPages, serviceRequestNumber, parentDocumentTitle
• Card catalog: titleOfWork, drawerName, timePeriod, cardIdentifiers[], cardImageUrls[] (scanned card images), feeDate
• displayNames — persons/organizations with roles (author, claimant)
• sourceUrl — official COINS detailed-record page, recordType, scrapedAt

Input

Field	Type	Default	Description
mode	string	search	search / searchByNumber / searchByOwner / searchByAuthor
searchQuery	string	harry potter	Free-text query (mode=search)
searchField	string	keyword	keyword (all fields) or title only
registrationNumber	string	–	Exact number lookup (mode=searchByNumber), e.g. TX0005739644, V3632D194
ownerQuery	string	–	Claimant/owner name (mode=searchByOwner)
authorQuery	string	–	Author name (mode=searchByAuthor)
workType	string	–	Type of work (literary, musical, visual arts, performing arts, sound recordings, motion pictures, software...)
registrationClass	string	–	Registration class (TX, PA, SR, VA, RE, MW, PR, ...)
registrationStatus	string	–	published / unpublished / unspecified
recordType	string	–	registration / recordation
recordationItemType	string	–	Assignment, mortgage/security agreement, DMCA filing, termination notice, licenses...
systemOfOrigin	string	–	voyager (post-1977) / card_catalog (pre-1978)
dateRangeFrom / dateRangeTo	string	–	Date range YYYY-MM-DD on the field in dateField
dateField	string	registration_date_as_date	Effective date of registration, publication, creation, certification, recordation, or representative date
containsKeyword	string	–	Only emit records whose title, claimants, authors or numbers contain this substring
sortBy	string	relevancy	Relevance / title / date / copyright number / latest transaction date
sortOrder	string	asc	asc / desc
maxResults	int	50	Hard cap (1–1000)
proxyConfiguration	object	–	Optional; only used automatically if the COINS API rate-limits or blocks direct requests

Example: search a famous title

{
  "mode": "search",
  "searchQuery": "harry potter",
  "maxResults": 5
}

Example: exact registration number lookup

{
  "mode": "searchByNumber",
  "registrationNumber": "TX0005739644"
}

Example: everything recorded by a studio, filtered by work type and date

{
  "mode": "searchByOwner",
  "ownerQuery": "Warner Brothers",
  "workType": "motion_picture",
  "dateRangeFrom": "2000-01-01",
  "dateRangeTo": "2010-12-31",
  "dateField": "registration_date_as_date",
  "maxResults": 100
}

Example: every book a specific author registered in a year

{
  "mode": "searchByAuthor",
  "authorQuery": "Rowling",
  "registrationClass": "TX",
  "dateRangeFrom": "1997-01-01",
  "dateRangeTo": "2007-12-31",
  "maxResults": 50
}

Use cases

• Licensing & rights research — confirm who owns a registered work before negotiating
• Publishing & entertainment legal teams — pull registration numbers, dates and renewal history for due-diligence
• Music / film libraries — verify sound-recording (SR) and performing-arts (PA) registrations and recordation documents
• Academic research — study registration trends by class, date and claimant across 60M+ records
• Card-catalog digitization projects — collect pre-1978 registration metadata plus scanned card images
• IP portfolio monitoring — track new registrations and assignments by claimant or author

Data Source

The US Copyright Office Public Records System ("COINS") at publicrecords.copyright.gov. This actor uses the same public JSON API the website itself calls (api.publicrecords.copyright.gov), so no scraping of rendered pages is required and no login is needed. Data is refreshed by the Copyright Office via its nightly export pipeline; every record carries sourceDate (export date) and a sourceUrl link to the official detail page for verification. This actor is not affiliated with or endorsed by the US Copyright Office.

Limitations

• No auth-walled axes: all modes (search, searchByNumber, searchByOwner, searchByAuthor) hit public endpoints — there is no private-registration or examiner-only data exposed.
• Card-catalog records are lighter: pre-1978 card-catalog records lack modern fields (claimant addresses, basis of claim, imprint) by design — the source has no such data; scanned card images are provided instead.
• Number formats are emitted as-is: registration-number display strings come straight from the Copyright Office (leading-zero quirks, I pub. 34656 style card-catalog numbers) to stay faithful to the official record.
• Search field limits: dateField supports the six date axes the public API exposes; the API does not index dates before 1938 (the card-catalog era uses representative_date).
• API stability: the public COINS API occasionally rate-limits or returns transient 5xx under heavy load; the actor retries with backoff and can use an optional proxy if direct access is blocked.
• Not affiliated: this actor is a third-party tool and is not affiliated with or endorsed by the US Copyright Office.

FAQ

Is the data free? Yes. The US Copyright Office publishes these public records free of charge; the actor uses the same public endpoint the COINS website uses.

What's the difference between a registration, a recordation and a card-catalog record? A registration is an original copyright claim filed with the Office (post-1978 records in the electronic catalog). A recordation is a document filed with the Office (assignment, security agreement, termination notice, license...). Card-catalog records are scanned pre-1978 registration cards. Records carry recordType so you can tell them apart.

Why are some registration numbers missing a leading zero? The Copyright Office's own display strings are emitted as-is (I pub. 34656 for card-catalog records, TX0005739644 style for modern ones). We also emit the parsed variants where the API provides them.

How do date filters work? dateRangeFrom/dateRangeTo are applied to the date axis you pick in dateField (default: effective date of registration). Use YYYY-MM-DD.

Does containsKeyword work on all fields? It filters on the record's title, variant titles, claimants, authors and copyright/registration numbers (case-insensitive substring).

How fresh is the data? The Copyright Office refreshes the public system roughly daily through its export pipeline; each record's sourceDate and sourceFile show which export it came from.

Can I get the official certificate? The actor returns the metadata and a link to the official detail page. Official certificates can be ordered from the Copyright Office directly.

Do I need a proxy? No. The public API is directly accessible. If the API rate-limits your run, you may optionally configure Apify proxy and it will be engaged automatically.