Open Library Scraper

Scrape Open Library — Internet Archive's open catalog of 50M+ books. Search by title / author / subject, fetch by ISBN or work ID, get full bibliographic metadata, cover images, edition counts, and full-text availability. HTTP-only via the public openlibrary.org JSON API. No auth, no proxy.

What this actor does

• Four modes: search, byIsbn, byWorkIds, byAuthorIds
• Universal IDs: OpenLibrary work IDs, ISBN-10, ISBN-13, author IDs
• Filters: publication year range, min edition count, language, full-text availability
• Cover images: auto-builds CDN URLs at S/M/L sizes
• Empty fields are omitted

Output per book

• workId — OpenLibrary work key (e.g. OL27482W)
• title, subtitle, description
• firstPublishYear, editionCount
• authors[], primaryAuthor, authorIds[]
• subjects[], subjectPlaces[], subjectPeople[], subjectTimes[]
• languages[] — 3-letter ISO codes
• isbn10[], isbn13[] — capped at 5 each
• coverUrl — large cover via covers.openlibrary.org CDN
• hasFulltext, publicScan, ebookAccess
• openLibraryUrl
• recordType: "book", scrapedAt

Input

Field	Type	Default	Description
mode	string	search	search / byIsbn / byWorkIds / byAuthorIds
searchQuery	string	the lord of the rings	Free-text query
queryAuthor	string	–	Constrain to author
querySubject	string	–	Constrain to subject
isbns	array	–	ISBNs (mode=byIsbn)
workIds	array	–	Work IDs (mode=byWorkIds)
authorIds	array	–	Author IDs (mode=byAuthorIds)
publishYearMin	int	–	Drop works before this year
publishYearMax	int	–	Drop works after this year
minEditionCount	int	–	Drop works with fewer editions
language	string	–	3-letter ISO (e.g. eng, fre)
includeFulltextOnly	bool	false	Only emit works with full-text on Internet Archive
maxItems	int	50	Hard cap (1–1000)

Example: classic fantasy with full-text scans

{
  "mode": "search",
  "searchQuery": "fantasy",
  "publishYearMin": 1900,
  "publishYearMax": 2000,
  "minEditionCount": 10,
  "includeFulltextOnly": true
}

Example: lookup by ISBN

{
  "mode": "byIsbn",
  "isbns": ["9780261103573", "9780261102217", "0-261-10357-X"]
}

Example: all of an author's works

{
  "mode": "byAuthorIds",
  "authorIds": ["OL26320A"],
  "maxItems": 200
}

Use cases

• Library systems — bulk-import metadata from Open Library by ISBN
• Edtech — discover books by subject for curriculum design
• Content discovery — find books by author / subject / time period
• Recommendation engines — feed Open Library subject taxonomy into your recommender
• Publishing intelligence — track edition counts to gauge popularity
• Academic research — bulk-export a subject's bibliographic record

FAQ

What's Open Library? An open, editable, library-grade catalog from Internet Archive. ~50M books, ~10M authors, free for any use. See openlibrary.org.

Is there a rate limit? Generous; no documented hard cap for normal scraping. The actor uses small delays to be polite.

What are work IDs vs edition IDs? A "work" is the abstract concept of a book (e.g. "The Hobbit"); an "edition" is a specific publication of it (paperback 1995 ed.). The actor returns work-level records by default.

Why are some ISBN lists capped at 5? Popular works have hundreds of editions, each with its own ISBN. We keep the first 5 for table compactness; full lists are available via per-edition queries.

How do subjects, subjectPlaces, subjectPeople, subjectTimes differ? Open Library's subject taxonomy is faceted: regular subjects (Fantasy), geographic places (Middle-earth), people (Bilbo Baggins), and time periods (Third Age).

What's ebookAccess? borrowable (Internet Archive lending), printdisabled (only for users with print disabilities), or no_ebook (text not digitized).

Can I get cover images at other sizes? The actor emits the L (large) URL. To get S or M, swap the suffix in the URL: -L.jpg → -M.jpg or -S.jpg.

How fresh is the data? Daily — Open Library re-indexes nightly from edits, ISBN ingestion, and Internet Archive's MARC import pipeline.