Apple Books MCP is a Python-based MCP server that exposes a set of tools for interacting with your Apple Books library. It provides capabilities such as listing collections and books, describing collections and books, retrieving annotations and highlights, searching notes and highlights, and performing full-text searches across annotations. The server is designed to be run locally or in a development environment and can be wired into Claude Desktop or other MCP clients via the standard MCP protocol. The available tools map to API-like operations such as list_collections, get_collection_books, describe_collection, list_all_books, get_book_annotations, describe_book, list_all_annotations, get_highlights_by_color, search_highlighted_text, search_notes, full_text_search, recent_annotations, and describe_annotation, enabling a rich set of interactions with your Apple Books data.

To use the server with Claude Desktop or an MCP client, configure an MCP entry (for example, apple-books-mcp) that runs the server locally. You can run the server via Python using the -m module invocation, which starts the Apple Books MCP service and exposes its endpoints to the MCP client. Once running, you can call the provided tools through the MCP protocol, pass parameters like collection_id or book_id as needed, and receive structured responses containing the requested data.

Prerequisites:

• Python 3.8+ installed on your system
• pip available to install Python packages
• Internet access to install the apple-books-mcp package

Installation steps:

1. (Optional) Create a virtual environment to isolate dependencies:

   python -m venv venv source venv/bin/activate # on macOS/Linux

   or

   .\venv\Scripts\activate # on Windows

2. Install the Apple Books MCP package from PyPI:

   pip install apple-books-mcp

3. After installation, run the MCP server using the specified entry point:

   python -m apple_books_mcp

4. Verify the server is running. If running in a development environment, you may connect via MCP clients once the server exposes its endpoints.

Notes and tips:

• The recommended way to run locally is via the Python module invocation (python -m apple_books_mcp). If you prefer a different launcher, you can configure uvx or other runtime environments as shown in the configuration examples.
• The Apple Books MCP server is currently documented with uvx and Python-based launch configurations. Docker support is planned but not yet available in the official release according to the README.
• When configuring Claude Desktop or other MCP clients, ensure the server name in your mcpServers map matches the actual server entry (e.g., apple-books-mcp) and that the command and arguments start the server correctly.
• If you modify any environment-related variables (e.g., library paths or Python virtual environment), reflect those changes in your MCP configuration or startup scripts to ensure proper operation.