Everlogue’s Reading Companion - a book recommendation agent grounded in structured Sanity
This is a submission for the Sanity Challenge, Path One: Ship an Agent That Queries Real Content What I Built Everlogue is a home for everything you read. It brings together a book catalog, private reading shelves, celebrity book-club selections, and Ask Everlogue , a Reading Companion grounded in structured Sanity content. The catalog began with selections from four clubs: Reese’s Book Club,…
Everlogue is a platform for book recommendations based on structured sanity content. The Reading Companion feature of Everlogue utilizes a curated book catalog and a knowledge base to provide personalized book suggestions. Initially, the catalog included selections from four book clubs: Reese's Book Club, Oprah's Book Club, Read with Jenna, and GMA Book Club.
Users can import their Goodreads lists and add books to their Everlogue shelves. If a book is not in the catalog, a request for review is created, and upon approval, the book joins the shared catalog.
The Reading Companion begins its search for new books by asking readers about their mood or favorite books. It then searches the full catalog for suitable books, excluding books already read or currently being read by the signed-in reader. The companion returns three recommendations with brief, personal, spoiler-free explanations, drawing upon actual books present in the user's Everlogue account.
For deeper exploration of a book, the Reading Companion accesses a Sanity Context Knowledge Base containing information from various sources such as book-club announcements, summaries, and author interviews. This knowledge base supports conversations about books, providing source-backed details on the book's premises, themes, and author conversations.
The Reading Companion uses two distinct content paths: the Knowledge Base for questions about books, and the structured catalog for recommendations. The Knowledge Base is accessed through Sanity Context and MCP (Managed Content Platform), while recommendations are made directly from the catalog using GROQ queries. Sanity stores relationships between books, genres, reader profiles, shelves, and feedback, allowing for personalized recommendations based on the reader's current mood and preferences.
The recommendation flow interprets the request, retrieves eligible books, and selects up to three, generating short explanations from the metadata provided. The companion's answers are grounded in retrieved sources and cited with their titles and URLs, ensuring accurate and reliable information.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
