Spines
Let an AI assistant search your Spines library, manage reading lists, and recognize books on a shelf photo while you research topics for a product or content project.
Overview
Spines MCP is a MCP server for the Idea phase that lets AI assistants access book collections, reading lists, and bookshelf recognition through the Spines API.
What is this MCP server?
- stdio MCP package @spines/mcp-server on npm (v1.0.2)
- Book collection access and reading-list management from the chat interface
- Bookshelf recognition from images for turning physical shelves into structured titles
- Requires SPINES_API_KEY secret environment variable
- Open-source server published from GetSpines/mcp-server on GitHub
- Package version 1.0.2
- stdio transport via @spines/mcp-server on npm
- Single required secret: SPINES_API_KEY
What problem does it solve?
Your research notes and reading lists live in Spines while your agent has no idea what you have already read or what is on your shelf.
Who is it for?
Solo builders and writers who already use Spines and want agent-native book context during early research.
Skip if: Teams needing full-text PDF RAG, non-book document libraries, or workflows with no Spines account.
What do I get? / Deliverables
After you set SPINES_API_KEY and register the stdio server, the assistant can list books, manage reading queues, and interpret shelf images as part of research chats.
- Agent-accessible book collection and reading-list operations
- Structured title candidates from bookshelf recognition workflows
Recommended MCP Servers
Journey fit
Reading lists and shelf recognition support discovery and source gathering before you commit to a build—classic Idea-phase research, not shipping or growth analytics. Book collections and curated lists are research inputs; integrations here belong on idea → research next to competitor and audience discovery tools.
How it compares
Spines library MCP bridge, not a general web-scraper skill or ebook storefront integration.
Common Questions / FAQ
Who is spines MCP for?
Indie researchers, authors, and builders who catalog books in Spines and want Claude Code or Cursor to reference those collections while exploring ideas.
When should I use spines MCP?
Use it in the Idea phase when curating reading lists, checking what you own, or digitizing a bookshelf photo before you lock product scope.
How do I add spines to my agent?
Install @spines/mcp-server from npm, configure stdio transport in your MCP settings, and set the required SPINES_API_KEY environment variable.