Books Scraper — Titles, Authors, ISBN, Editions (Open Library)

Search 20+ million Open Library book records by title, author, subject, publisher or ISBN and get clean JSON: authors, first publish year, edition count, ISBNs, subjects, languages, publishers, ratings, reading-log popularity, Goodreads/Amazon/OCLC ids, covers and e-book availability. Look up ISBNs for publisher, pages, format and description; list an author's works, browse a subject, or pull every edition of a work. No API key, no proxy, no browser.

Use cases

• Catalogue enrichment for bookshops and marketplaces: turn a list of ISBNs into title, authors, publisher, pages, format, cover and description.
• Publisher and competitor research: list a publisher's catalogue (searchField: "publisher") or a subject's newest titles.
• Rights and translation scouting: find which languages and editions exist for a work, with years and publishers.
• Reading-list and recommendation products: rank books in a subject by popularity (reading logs, want-to-read) or rating.
• Author due diligence: bio, birth/death dates, Wikidata/VIAF/ISNI ids and most-read works of an author.
• Library and e-learning: filter to works with a readable/borrowable scan on archive.org (ebookOnly).

Input

Field	Type	Default	Allowed values / notes
queries	string[]	required	Titles, keywords, author names, subjects, ISBN-10/13 (hyphens ok), Open Library keys OL…W / OL…M / OL…A or their URLs
mode	string	auto	auto (detect ISBN/key/text), search, isbn, author, subject, work
searchField	string	any	any, title, author, subject, publisher, place, person, isbn (search mode)
limit	integer	5	1–100 results per query (search/author/subject)
offset	integer	0	0–10 000, for paging across runs
sortBy	string	relevance	see Sort orders; subject mode: only relevance/editions (most editions), new, old — other sorts are refused there
language	string	any	MARC 21 code, see Languages; search and author modes (refused in subject mode: the subjects API has no language filter)
yearFrom / yearTo	integer	none	1000–2100, first publication year (inclusive)
ebookOnly	boolean	false	Only works with a scan on archive.org
includeAvailability	boolean	false	Lending status per work (search mode)
includeDescription	boolean	false	Work description + external links (+1 request/work; ISBN, edition and work lookups always carry the description)
includeEditions	boolean	false	Editions list per work (+1 request/work)
maxEditions	integer	10	1–50
includeAuthorDetails	boolean	false	authorsDetailed: bio, dates, Wikipedia/Wikidata/VIAF/ISNI, photo (+1 request per distinct author, cached)
coverSize	string	L	S, M, L
maxSubjects	integer	15	0–200
dedupe	boolean	true	Save each work/edition/author once across queries
fields	string[]	all	Keep only these output fields, in your order; query is always kept. Case, snake_case and Open Library names (author_name, cover_i) are read as the field; unknown names are left out and named in the status; a list with no known field is refused. Rows with none of the listed fields (e.g. the author row for ["title"]) are not saved and not charged

Reference

Lookup modes

Mode	What you pass	What you get
auto	anything	ISBN → edition record; OL…W → work; OL…M → edition; OL…A → author + works; text → search
search	keywords	Works matching the text (optionally in one searchField)
isbn	ISBN-10/13	One edition: publisher, date, pages, format, identifiers, work description, subjects, author names. Check digit is validated (warning only)
author	name or OL…A	One author row (for names, the record with the most works is chosen, alternatives are logged) + their works
subject	subject name/slug	Works in that subject (science fiction → science_fiction), with total subjectWorkCount. Years and ebookOnly apply; language does not exist there
work	OL…W / OL…M	Full work record from the search index + description

Sort orders

Value	Meaning
relevance	Best text match (default)
new / old	First publish year, newest / oldest first
rating	Average user rating
editions	Most editions
readinglog	Most reading-log entries (overall popularity)
want_to_read / currently_reading / already_read	Reading-log shelves
title	Title A–Z
scans	Works with scans first
random	Random order

Languages

The language select offers these MARC 21 codes; Apify refuses any other value before the run starts. Full MARC list: loc.gov MARC language codes.

Code	Language	Code	Language	Code	Language	Code	Language
eng	English	spa	Spanish	fre	French	ger	German
ita	Italian	por	Portuguese	rus	Russian	chi	Chinese
jpn	Japanese	kor	Korean	ara	Arabic	dut	Dutch
swe	Swedish	pol	Polish	tur	Turkish	per	Persian
heb	Hebrew	hin	Hindi	ukr	Ukrainian	cze	Czech
dan	Danish	fin	Finnish	nor	Norwegian	hun	Hungarian
gre	Greek	rum	Romanian	kaz	Kazakh	uzb	Uzbek
kir	Kyrgyz	tgk	Tajik	aze	Azerbaijani	geo	Georgian
arm	Armenian	mon	Mongolian	bel	Belarusian	tat	Tatar
ind	Indonesian	may	Malay	vie	Vietnamese	tha	Thai
ben	Bengali	urd	Urdu	tam	Tamil	tel	Telugu
mar	Marathi	lat	Latin	cat	Catalan	bul	Bulgarian
hrv	Croatian	srp	Serbian	slo	Slovak	slv	Slovenian
lit	Lithuanian	lav	Latvian	est	Estonian	ice	Icelandic
afr	Afrikaans	swa	Swahili	tgl	Tagalog

Common subject slugs

Any Open Library subject works; examples: business, entrepreneurship, management, economics, finance, marketing, leadership, self-help, psychology, science_fiction, fantasy, mystery_and_detective_stories, romance, history, biography, programming, computer_science, artificial_intelligence, machine_learning, mathematics, medicine, cooking, travel, children, poetry. Browse more at openlibrary.org/subjects.

Examples

Enrich a list of ISBNs for a shop catalogue

{ "queries": ["978-0-307-88789-4", "9780062316097", "0735211299"], "mode": "isbn", "includeAuthorDetails": true, "coverSize": "M" }

Newest books in a subject (new-release tracking)

{ "queries": ["entrepreneurship"], "mode": "subject", "yearFrom": 2024, "sortBy": "new", "limit": 50 }

An author's most-read works with all editions (rights scouting)

{ "queries": ["Peter Drucker"], "mode": "author", "sortBy": "readinglog", "limit": 10, "includeEditions": true, "maxEditions": 20, "includeAuthorDetails": true }

A publisher's catalogue in one language

{ "queries": ["O'Reilly Media"], "searchField": "publisher", "language": "eng", "yearFrom": 2020, "sortBy": "new", "limit": 100, "fields": ["title", "authors", "firstPublishYear", "isbn13", "url"] }

Free-to-read classics for an e-learning library

{ "queries": ["economics"], "searchField": "subject", "ebookOnly": true, "includeAvailability": true, "sortBy": "readinglog", "limit": 25 }

Output

Search / author / subject rows describe a work; ISBN rows describe an edition; author mode adds one author row. Real example (author-mode work row for "Peter Drucker", sortBy: "readinglog"; arrays trimmed):

{
  "query": "Peter Drucker",
  "mode": "author",
  "workKey": "OL274337W",
  "title": "The Practice of Management",
  "authors": ["Peter F. Drucker"],
  "authorKeys": ["OL4345625A"],
  "firstPublishYear": 1954,
  "latestPublishYear": 2017,
  "editionCount": 37,
  "isbn13": "9780060913168",
  "isbn10": "0330020315",
  "isbnCount": 44,
  "languages": ["chi", "und", "eng"],
  "publishers": ["Butterworth Heinemann", "Taylor & Francis Group"],
  "subjects": ["Business", "Nonfiction", "Industrial management"],
  "pagesMedian": 404,
  "firstSentence": "THE manager is the dynamic, life-giving element in every business.",
  "rating": 5,
  "ratingsCount": 2,
  "readingLogCount": 144,
  "wantToRead": 130,
  "goodreadsId": "114837",
  "amazonAsin": "B002JN0B82",
  "oclc": "567448468",
  "ddc": "658",
  "ebookAccess": "borrowable",
  "hasFulltext": true,
  "coverUrl": "https://covers.openlibrary.org/b/id/6418448-L.jpg",
  "url": "https://openlibrary.org/works/OL274337W",
  "source": "openlibrary",
  "fetchedAt": "2026-09-13T08:17:11.726Z",
  "numFound": 320
}

Field	Rows	Description
query, mode, found	all	Input query (always kept, also with fields), lookup type; found is true on every row (queries that match nothing or fail are not rows — see SUMMARY)
workKey, editionKey, authorKey	work / edition / author	Stable Open Library ids
title, subtitle, authors, authorKeys	work, edition
firstPublishYear, publishYears, latestPublishYear, editionCount	work
isbn13, isbn10, isbns, isbnCount	work (strings: the first ISBN-13/-10; isbns: first 20), edition (lists)
languages, publishers, subjects, places, people, timePeriods	work
rating, ratingsCount, readingLogCount, wantToRead, currentlyReading, alreadyRead	work	Popularity signals
goodreadsId, amazonAsin, librarythingId, oclc, lccn, lcc, ddc	work, edition	External identifiers / classifications
ebookAccess, hasFulltext, publicScan, archiveOrgIds, availability	work	E-book status
publishDate, publishYear, pages, physicalFormat, publishPlaces, series, edition, description	edition
editions, editionsTotal	with includeEditions	Brief edition list: key, year, publishers, ISBNs, languages, pages, format, cover
authorsDetailed, name, bio, birthDate, deathDate, wikidataId, viaf, isni, photoUrl	author / includeAuthorDetails	The dataset's Authors view shows the author rows
numFound, subjectWorkCount	search / subject	Total matches at the source
coverUrl, url, source, fetchedAt	all

Every field's type, meaning and an example are in the dataset schema (Output tab / API).

Use it from code / agents

curl -X POST "https://api.apify.com/v2/acts/yadroo~openlibrary-books/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"queries":["9780307887894","9780062316097"],"mode":"isbn"}'

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('yadroo/openlibrary-books').call({ queries: ['machine learning'], mode: 'subject', sortBy: 'new', limit: 30 });
const { items } = await client.dataset(run.defaultDatasetId).listItems();

from apify_client import ApifyClient
client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("yadroo/openlibrary-books").call(run_input={"queries": ["Nassim Taleb"], "mode": "author", "includeAuthorDetails": True})
items = client.dataset(run["defaultDatasetId"]).list_items().items

MCP: add https://mcp.apify.com to Claude / Cursor / any MCP client and call the yadroo/openlibrary-books tool with the same JSON input.

Pricing

Pay per event: $0.001 per run start + $0.001 per dataset row (work, edition or author). Queries that match nothing or fail are not rows and are not charged. Store discounts: Bronze −10 %, Silver −20 %, Gold and above −30 % on the row price; the start event is the same on every plan; platform usage is included. Typical runs: 10 ISBN lookups ≈ $0.011; 100 books in a subject ≈ $0.101; an author row + 50 works ≈ $0.052. The run never goes over your Maximum cost per run: it stops with the status Stopped at your spending limit: N rows delivered, and it spends enrichment requests only on rows the limit still pays for.

Limits & FAQ

• Speed and politeness: Open Library is a volunteer-run non-profit. Its API policy allows 3 requests per second to clients that identify themselves, as this actor does with its User-Agent. The actor sends one request at a time, at least 350 ms apart (retries included). An ISBN costs about 3 requests: the edition, its work (subjects, description) and its author names, with authors cached within the run. Enrichment options add one request per work, and includeAuthorDetails adds one per distinct author.
• Capacity: a 15-minute run (the default timeout) does roughly 2 000 requests. That is about 600–700 ISBNs, or a few hundred enriched works. Near the timeout the actor stops starting requests, saves what it has and ends with the status Stopped before the run timeout: N rows saved. Raise the timeout or split the job for more.
• Bulk data: Open Library asks not to bulk-download through its API. For whole-catalogue work (millions of records) use its free monthly data dumps instead of this actor.
• Data quality: records are community-edited: duplicates of authors/works exist, ratings are sparse, some ISBNs map to the wrong work. For name lookups in author mode the record with the most works is chosen and alternatives are logged; pass an OL…A key to force one.
• Freshness: live API, new records appear as soon as they are added to Open Library.
• Not found / errors: ISBNs, keys or searches that match nothing, and requests that fail, are not dataset rows (you are not charged for them). They are listed in the run's SUMMARY record (notFound, notFoundCount, errors: [{ input, error }], errorCount) and in the status message, e.g. 7 record(s) saved · 1 query not found · 1 failed. The run fails only when nothing could be fetched at all. Each request gets 30 s per attempt and up to 3 attempts. Retries on 429/5xx, dropped connections and non-JSON answers wait 2 s and then 4 s, or whatever Retry-After asks for. When 3 requests in a row fail like that, Open Library is treated as down: the run stops and the status names the queries not tried.
• Not included: prices, stock or seller offers (Open Library does not have them), and full text of books.
• Roadmap: trending lists.

---

Made by Yadroo. Sibling actors: arxiv-papers, openalex-works, wikipedia-search, rss-to-json, hackernews-search.