Emby.MCP

Model Context Protocol (MCP) server that connects an Emby media server to an AI client such as Claude Desktop, creating a kind of Amazon Alexa ™ for any MCP compliant LLM using your own media collection. A personal MCP and Python learning exercise, inspired into action by hearing Yoko Li talk about her Morse Code MCP server.

Note, this is an independent project with no affiliation to, or endorsement by, Emby LLC.

Contents

Features | Requirements | Installation | Usage | Under The Hood | Also See | License

Features

A minimum viable project that allows an LLM to do the following via MCP tools:

• Log on to & log out of an Emby media server;
• Retrieve a list of media libraries;
• Select a specified library;
• Retrieve a list of genres used in that library;
• Search for items in that library by genre, item title, album name, release year, and lyrics, chunking the response as needed;
• Retrieve playlists, create new playlists, add items to playlists & re-order them, and share playlists with other Emby users;
• Retrieve a list of accessible media players known to Emby;
• Retrieve the current play queue of a specified media player;
• Control the playing, pausing, seeking, etc of a specified media player, including transferring the queue to another player.

Requirements

• Python v3.13 or higher
• The uv package/project manager
• The MCP Server SDK for Python v1.94 or higher
• The Emby client SDK for Python v4.9.0.33, with hotfix patches (see below).
• A working Emby Media Server with a library of media files, either local or accessible via a network.
• An MCP-compatible LLM/AI client. Emby.MCP was developed using Claude Desktop, however at time of writing more than 60 other clients were listed in the Feature Support Matrix - pick one that supports Tools.
• An LLM subscription plan that supports MCP (this may require payment).

Installation

The following instructions are for a clean installation on Windows 11 Pro - adjust for your own platform and needs.

Install Python | Install Emby.MCP | Install Patches | Login Config | Basic Checks | Configure LLM

Install Python

• Install latest Python. Customise the install: Optional Features = select all | Advanced Options = install for all, associate files, create shortcuts, add to environment, precompile.
• From a Powershell terminal, run:

pip install uv