MCP Open Library

What 8enSmith/mcp-open-library does

The 8enSmith/mcp-open-library MCP server connects an MCP-compatible assistant to the Open Library catalogue. It provides read-only operations for discovering books and authors, looking up detailed records, and obtaining image URLs.

Book searches can combine fields including free-form queries, titles, authors, subjects, places, people, publishers, and ISBNs. Author searches accept a name, while author details require an Open Library author key. Book details can be retrieved with identifiers such as ISBN, LCCN, OCLC, or OLID.

How it works

The assistant calls one of the exposed MCP tools, and the server requests data from Open Library. Responses are returned as JSON projections rather than unfiltered API payloads. Search responses include the number of matches, the current offset, the requested limit, and result records.

Search tools return up to 50 results per call and use an offset for paging, with an offset limit of 1000. A book search must include at least one filtering field; an unfiltered catalogue request is rejected. The q parameter can contain a free-form Solr query, and optional parameters support language codes and sorting.

Search results identify works but also include a best_edition record with an edition key. That OLID can be passed to get_book_by_id to retrieve a specific edition and its available ISBN arrays. Tool failures, including rejected arguments or an unreachable Open Library API, are returned as error-marked tool results.

Setup and configuration

The 8enSmith/mcp-open-library MCP server can run from the published npm package without a separate build step. Add the following server entry to an MCP client configuration:

{
  "mcpServers": {
    "mcp-open-library": {
      "command": "npx",
      "args": ["-y", "mcp-open-library"]
    }
  }
}

For Claude Desktop, place the entry in claude_desktop_config.json and restart the client. Manual setup is also documented through repository cloning, npm install, and npm run build. The documented development environment uses Node.js version 22.21.1, though the project notes that newer versions may work.

Tools and capabilities

The server exposes seven tools:

• search_books searches by combinations of catalogue fields and supports sorting and paging.
• get_book_by_title searches for books using a title.
• get_authors_by_name finds authors by name with paging.
• get_author_info retrieves detailed data from an Open Library author key.
• get_author_photo returns an author photo URL from an Open Library ID.
• get_book_cover returns a cover URL using ISBN, OCLC, LCCN, OLID, or another supported book identifier.
• get_book_by_id retrieves detailed book data using ISBN, LCCN, OCLC, or OLID.

The cover and author-photo operations verify that an image exists before returning its URL. All tools advertise read-only and open-world annotations, but the MCP client determines whether those hints affect confirmation prompts.

Limitations and notes

The server depends on the Open Library API being reachable and reflects the identifiers and data available there. Search result ISBN fields may be absent when an edition has no recorded ISBN; when multiple ISBNs exist, search projections report the first of a type, while detailed edition lookups return all available values. The server does not modify books, authors, editions, or catalogue records.

Getting started with this 8enSmith/mcp-open-library MCP server

Always refer to the official documentation for the most accurate and up-to-date information.