Books

Book tools draw on Open Library and Google Books. Neither needs an API key; a Google Books key raises its rate limit.

search_books

Search Books. Read-only. Reaches outside the server.

Search for books by title, author, or free text.

Uses Open Library. Prefer title and author over query: they use dedicated indexes and return far better results.

Open Library is politeness-rate-limited, so a busy moment may answer with a job handle to poll using get_job_result instead of the results themselves.

Parameters

Name	Type	Default	Description
query	string \| null	null	Free-text fallback. Use title/author when known.
title	string \| null	null	Book title or partial title (recommended).
author	string \| null	null	Author name (recommended).
limit	integer	10	Maximum results to return (max 50).

Returns

{"books": [...]}: records with title, authors, publisher, year, ISBNs, Open Library IDs, cover URL, and subjects.

At least one of query, title, or author must be provided.

When only query is given, it is first tried as a title search (better relevance) and falls back to free-text if no results are found. When author is given with multiple tokens (such as "Frank Duffy") and initial results are thin, a broadened search is automatically attempted to catch name variants (such as Frank → Francis).

A response looks like:

{
  "books": [
    {
      "title": "Deep Learning",
      "authors": [
        "Ian Goodfellow",
        "Yoshua Bengio",
        "Aaron Courville"
      ],
      "publisher": "MIT Press",
      "year": 2016,
      "edition": null,
      "isbn_10": "0262035618",
      "isbn_13": "9780262035613",
      "openlibrary_work_id": "OL17953442W",
      "openlibrary_edition_id": "OL26423929M",
      "cover_url": "https://covers.openlibrary.org/b/isbn/9780262035613-M.jpg",
      "google_books_url": "https://books.google.com/books?id=Np9SDQAAQBAJ",
      "worldcat_url": "https://www.worldcat.org/isbn/9780262035613",
      "snippet": "An introduction to a broad range of topics in deep learning...",
      "cover_path": null,
      "subjects": [
        "Machine learning",
        "Artificial intelligence"
      ],
      "page_count": 800,
      "description": null
    }
  ]
}

get_book

Get Book. Read-only. Reaches outside the server.

Fetch book metadata by ISBN or Open Library ID.

Open Library is politeness-rate-limited, so a busy moment may answer with a job handle to poll using get_job_result instead of the record itself.

Parameters

Name	Type	Default	Description
identifier	string	required	ISBN-10, ISBN-13, Open Library work ID (such as OL1168083W), or edition ID (such as OL1429049M).
download_cover	boolean	false	If True, download and cache the cover image locally. Returns cover_path in the response. In read-only mode, returns cover_error instead.
cover_size	string	"M"	Cover size variant: "S" (small), "M" (medium), "L" (large). Defaults to "M".

Returns

The book record, or {"error": "not_found"} if not found.

Identifier formats:

Format	Example
ISBN-13	9780262035613
ISBN-10	0262035618
ISBN with hyphens	978-0-262-03561-3
Open Library work ID	OL17953442W
Open Library edition ID	OL26423929M

The book record has the same shape as items returned by search_books, and the not_found error also carries "identifier": "...".

Results are cached. Work and edition lookups are cached by their respective Open Library IDs; ISBN lookups are also stored under the resolved ISBN-13. When download_cover=true, the cover image is saved to <cache_dir>/covers/<isbn>_<size>.jpg.

get_book_excerpt

Get Book Excerpt. Read-only. Reaches outside the server.

Get a book excerpt and preview info from Google Books.

Returns the publisher description, text snippet, and a link to the Google Books preview page. Google Books does not expose full chapter text through its API; the excerpt is a publisher-provided summary and/or search snippet.

An error of not_found means Google Books answered and has no such volume: stop asking about that ISBN. A lookup that never got an answer says so instead, as rate_limited (with retryable) or upstream_error, and is worth retrying: the book may well exist.

Parameters

Name	Type	Default	Description
isbn	string	required	ISBN-10 or ISBN-13.

Returns

JSON with excerpt, description, preview availability, and link.

Example return::

{"excerpt": "...", "description": "...", "source": "`google_books`",
 "`preview_available`": true, "`preview_link`": "https://..."}

A response looks like:

{
  "excerpt": "Deep learning is a form of machine learning that enables...",
  "description": "An introduction to a broad range of topics in deep learning...",
  "source": "google_books",
  "preview_available": true,
  "preview_link": "https://books.google.com/books?id=Np9SDQAAQBAJ&printsec=frontcover"
}

Error shapes

not_found is {"error": "not_found", "isbn": "..."}. A lookup that never got an answer is {"error": "rate_limited", "isbn": "...", "retryable": true} for a 429, or {"error": "upstream_error", "isbn": "...", "status": <code or null>} for anything else.

Without SCHOLAR_MCP_GOOGLE_BOOKS_API_KEY set, requests use the anonymous tier, where 429 is common.

recommend_books

Recommend Books. Read-only. Reaches outside the server.

Recommend books for a subject via Open Library.

Uses the Open Library subject API to find popular books on a topic, sorted by edition count (a proxy for popularity).

Parameters

Name	Type	Default	Description
subject	string	required	Subject or topic, such as machine learning, algorithms or computer vision.
limit	integer	10	Maximum results to return (max 50).

Returns

JSON list of book records sorted by popularity.

Each record has the same shape as search_books results, with fields populated from the Open Library subject API (title, authors, Open Library work ID, cover URL). ISBN and edition fields are null since subject results are work-level.

Results are cached for 7 days, keyed by the normalized subject slug. Up to 50 results are fetched and cached; the limit parameter slices the cached pool on return.