spotify-llm-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| now_playingA | Get the current playback state: track, artist, device, progress, shuffle/repeat. |
| controlB | Control playback. action ∈ {play, pause, play_pause, next, previous, shuffle, repeat, volume, seek}. For 'volume' pass 0-100; for 'seek' pass milliseconds. |
| playA | Search Spotify and play the top matching track. |
| play_playlistB | Start playback of a playlist by its Spotify id (optionally shuffled). |
| searchA | Search Spotify tracks; returns up to |
| list_devicesA | List available Spotify Connect devices. |
| switch_deviceB | Transfer playback to the device with the given name. |
| get_queueB | Get the current playback queue. |
| list_playlistsA | List the user's playlists (id, name, description, owner). |
| get_playlist_tracksA | Get a playlist's tracks by name (case-insensitive) or by Spotify id. Use this to read a playlist before recommending music in its vibe. |
| taste_profileA | Get the user's top tracks, to reason about their taste. |
| resolve_tracksA | Resolve a list of {title, artist} picks to real Spotify tracks. Returns {resolved: [...], unresolved: [...]}. Use to preview before building. |
| create_playlist_from_tracksA | Create a playlist from LLM-proposed {title, artist} picks. The server searches each on Spotify, adds the confident matches, and reports any it couldn't find under 'unresolved'. Returns the new playlist_id + uri. |
| append_to_playlistA | Append LLM-proposed {title, artist} picks to an existing playlist by id. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| recommend_from_playlist | Extend a playlist with more music in its vibe, then play it. |
| vibe_playlist | Build a fresh playlist matching a described mood/vibe. |
| more_like_this | Recommend and queue music similar to what's playing now. |
| my_taste | Summarize the user's listening taste. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 14 tools
Most tools have a clear resource+action distinction, but search, play, and resolve_tracks overlap in their underlying lookup/resolution behavior, and the playlist creation/append tools both resolve track picks. Descriptions are usually enough to pick the right one.
Names are mostly readable but inconsistent: verb_noun forms like play_playlist and list_devices coexist with bare verbs like play and control, plus noun phrases like now_playing and taste_profile. The generic name control is particularly vague.
14 tools is a reasonable size for a Spotify assistant covering playback, queue, devices, search, and playlist management. Some search/resolve and playback-control functionality could be consolidated, but the count is not excessive.
Core workflows are covered: playback control, queue, devices, search, taste profile, and creating/appending playlists. However, there is no way to remove tracks or delete/replace playlists, nor any queue editing or recommendation tool, leaving notable playlist-lifecycle gaps.