Choose an endpoint

Pick one call. Do not search when you already have an ISBN or an id.

You have	REST	MCP	Why
ISBN-10 or ISBN-13	GET /v1/isbn/:isbn	lookup_isbn	Exact, cached, cheapest
Edition id fds_ed_…	GET /v1/books/:id	get_book	Point lookup
Work id fds_wk_…	GET /v1/works/:id	get_work	Synopsis lives on the work
Author id fds_au_…	GET /v1/authors/:id	—	Contributor record
A title to identify the book	POST /v1/match	resolve_book	Ranked candidates; take the first
An author name only	GET /v1/authors/search	search_authors	Not search_books
Genre browse	GET /v1/books/search?genre=	search_books + genre=	Closed facet, not q=epic fantasy
Books like these	POST /v1/recommend	recommend_books	Content similarity, not social taste
Exploratory / filtered browse	GET /v1/books/search	search_books	Not the first choice for a known title
Sync since a cursor	GET /v1/changes	get_changes	Empty on this first release
Parquet dump	GET /v1/datasets/books/releases/latest	—	Same Bearer key

Decision rules

1. ISBN present → lookup_isbn / GET /v1/isbn/:isbn. Never pass an ISBN as q on search.
2. Stable id present → the matching GET /v1/{books|works|authors}/:id.
3. User named a book (“the Andy Weir Hail Mary”) → resolve_book / POST /v1/match. Read candidates[0].
4. Author only → search_authors / GET /v1/authors/search.
5. Genre browse → search_books with genre=fantasy (not q= as a genre name).
6. Books like these → recommend_books / POST /v1/recommend. Content similarity.
7. User is browsing titles → search_books with filters.
8. You are syncing a cache → get_changes. On release 2026-09.2 the list is empty.

Ranking note

Search and match fold articles/apostrophes/subtitles, drop study guides and companions when a real title exists, prefer primary over graphic novels unless the query looks graphic, then collapse to one canonical edition for that work.

Check record

9780593135204 / Project Hail Mary / Andy Weir. Expect a description and a non-empty authors array.