Applemusic Mcp

Apple Music MCP server: playlists, library, catalog, playback and the Up Next queue, on macOS, Windows and Linux. A Mac or an Apple Music subscription is all yo…

LLM Mart 0 views 1 listing impressions
Transport
Not stated
Package
—
Registry id
—

No install snippet on purpose. A working MCP config is a command, its arguments and an environment block — the last two are where API keys live, so this catalogue never stores them and cannot publish them. Follow the link above for the authors' own instructions.

MCP server for Apple Music. It gives any MCP client (Claude, Cursor, Cline, Windsurf) control of your playlists, library, catalog, discovery, playback, and the Up Next queue. Runs on macOS, Windows, and Linux.

A Mac or an Apple Music subscription is all you need. Four engines, chosen per call: Music.app and Safari on a Mac, the Apple Music API and Chrome on any OS.

Features

Four engines back the server. Native drives the local Music.app on macOS via AppleScript. API uses Apple Music's web API (amp-api.music.apple.com) on any OS. Safari drives your signed-in Safari's MusicKit on macOS (DRM-native, zero install). Chrome runs a local Google Chrome window with MusicKit for DRM audio on any OS. One mode preference picks the engine — auto (default) mixes the best of each: native Music.app for playback on macOS, Safari for the Up Next queue, the API for data, Chrome off-mac. Pin one with native / safari / chrome / api, or override a single playback/queue call with engine=. In the table below, the Browser column covers both the Safari and Chrome web players. ✓ supported, ✗ not possible on that engine, — not applicable there.

Capability Native (Music.app) macOS API (amp-api) any OS Browser (Safari macOS / Chrome any OS)
Catalog search / browse ✓ ✓ (+ tokenless resolve) —
Recommendations / charts / suggestions ✗ ✓ —
Library search / browse ✓ ✓ —
Genre search ✓ ✗ —
Recently played / added ✓ ✓ —
Add catalog → library ✓ ✓ ✓ (in-page POST)
Remove from library ✓ ✓ —
Love / dislike ✓ ✓ —
1–5 star ratings ✓ ✗ ✗
Favorites list ✓ ✗ ✗
Playlist create / add / remove / rename ✓ ✓ —
Playlist copy ✓ ✗ —
Playlist delete ✓ ✓ (web token) —
Folders: single level + move in/out ✓ ✓ —
Folders: nested paths / tree / path ✓ ✗ ✗
Playback: play song / album / playlist / URL ✓ — ✓
Controls: pause / stop / next / prev / seek ✓ — ✓
Settings: volume / shuffle / repeat ✓ — ✓
now_playing ✓ — ✓
Up Next queue: view / next / last / remove / jump / clear / autoplay ✗ — ✓
Reveal in app ✓ — ✓ (navigates page)
AirPlay device select ✓ ✗ ✗
Library snapshot / integrity ✓ ✗ ✗
Works with no Apple account ✓ ✗ ✗
Cross-platform (Win/Linux) ✗ ✓ ✓

Everything in the API column runs anywhere, no browser and no Music app. Browser playback and the queue need a desktop session and a web player — Safari on macOS (no install) or Google Chrome elsewhere.

Setup

Requirements: Python 3.10+, plus either a Mac or an Apple Music subscription. The Chrome web player (cross-platform playback + Up Next queue) needs Google Chrome + Playwright. On macOS you can skip both — sign in via Safari and play through the Music app — so the default macOS install is lightweight (no ~500 MB Playwright). Windows/Linux include Playwright automatically (it's the only path there).

Claude Code, one line:

claude mcp add "Apple Music" -- uvx applemusic-mcp serve

Claude Desktop / Cursor / Cline / Windsurf, install once, then add the config block:

pipx install applemusic-mcp        # or: pip install applemusic-mcp
# Off macOS, also fetch the browser engine:  playwright install chromium
# macOS needs neither (Safari sign-in + Music.app). For the Chrome web player on a
# Mac:  pipx install 'applemusic-mcp[browser]'  then  playwright install chromium
{
  "mcpServers": {
    "Apple Music": {
      "command": "applemusic-mcp",
      "args": ["serve"]
    }
  }
}

Restart your client and try "List my Apple Music playlists" or "Play my favorites." On macOS the local library and playback work immediately. To add catalog music or run on any OS, sign in.

Sign in

Two paths capture the credentials for the cross-platform API.

From the project's README.

Related servers

Let LLM help you achieve your regression analysis with Stata

18 views

vSphere with Tanzu (VKS): Namespace and TanzuKubernetesCluster lifecycle. Requires vSphere 8.x+.

17 views

Web search, deep research, prediction markets & crypto data for AI agents. Pay per call via x402.

16 views

Durable agentic memory, encrypted at rest. Fully offline: no network, no API key, no cloud.

15 views