spotifyify 0.8.0

Async wrapper for spotipy with a focus on integration with MCP agents.

spotifyify

An async-first Spotify client with a namespaced API and fully typed response models maintained in the codebase and aligned with the official Spotify OpenAPI specification.

Requirements

• Python 3.13+
• A Spotify app registered at developer.spotify.com

Installation

pip install spotifyify

With uv:

uv add spotifyify

Optional CLI:

uv add "spotifyify[cli]"
spotifyify --help

See CLI for global installation and usage examples.

Configuration

Credentials are loaded from environment variables (or a .env file via pydantic-settings):

Variable	Required	Description
SPOTIFY_CLIENT_ID	Yes	Your app's client ID
SPOTIFY_CLIENT_SECRET	Yes	Your app's client secret
SPOTIFY_REDIRECT_URI	For user auth	OAuth redirect URI
SPOTIFY_ACCESS_TOKEN	Optional	Pre-existing access token
SPOTIFY_REFRESH_TOKEN	Optional	Refresh token for automatic renewal
SPOTIFY_TOKEN_EXPIRES_AT	Optional	Unix timestamp when the token expires

.env example:

SPOTIFY_CLIENT_ID=your_client_id
SPOTIFY_CLIENT_SECRET=your_client_secret
SPOTIFY_REDIRECT_URI=http://localhost:8888/callback
SPOTIFY_REFRESH_TOKEN=your_refresh_token

If SPOTIFY_REFRESH_TOKEN (or SPOTIFY_ACCESS_TOKEN) is present the client operates in user mode. Without them it falls back to the Client Credentials flow, which only allows access to public data.

If a user-scoped endpoint is called and no user token is available, spotifyify starts an interactive Authorization Code login:

1. Opens the Spotify consent page in your browser.
2. Waits for the redirect on your configured SPOTIFY_REDIRECT_URI (for example http://localhost:8888/callback).
3. Exchanges the code for access and refresh tokens.
4. Stores tokens in .spotify_cache by default (already git-ignored in this project).

Bring your own user token

Multi-user backends can pass an already minted end-user access token for one request scope. While the context is active, spotifyify sends that token directly and skips its configured OAuth provider, app credentials, cache, refresh, and scope checks. Expired or insufficient tokens surface as Spotify API errors.

The token context is isolated per async task, so one shared Spotifyify instance can serve concurrent users:

async with Spotifyify() as spotify:
    async with spotify.session(access_token=end_user_access_token):
        playlist = await spotify.playlists.create("Weekly Mix", public=False)
        await spotify.playlists.replace(
            playlist.id,
            ["spotify:track:...", "spotify:track:..."],
        )

Quick start

import asyncio
from spotifyify import Spotifyify, SpotifyScope

async def main():
    async with Spotifyify(scopes=[SpotifyScope.USER_READ_PLAYBACK_STATE]) as sp:
        state = await sp.player.state()
        if state and state.item:
            print(f"Now playing: {state.item.name}")

asyncio.run(main())

API design

The Spotifyify class is the entry point. It exposes all Spotify resources as lazy-loaded namespace properties. Every method is a coroutine and must be awaited.

Spotifyify
├── .tracks      # Tracks, search
├── .artists     # Artists, top tracks, discography, related artists
├── .albums      # Albums, new releases
├── .playlists   # Playlists, CRUD, track management
├── .player      # Playback control, queue, devices, history
├── .library     # Saved tracks/albums/shows/episodes, top items
├── .shows       # Podcast shows
├── .episodes    # Podcast episodes
└── .users       # Current user, public profiles, following

Tracks — sp.tracks

Method	Description
find(query, *, limit, offset, market)	Search for tracks
get(track_id, *, market)	Get a single track
get_many(track_ids, *, market)	Get up to 50 tracks

Artists — sp.artists

Method	Description
find(query, *, limit, offset)	Search for artists
get(artist_id)	Get a single artist
get_many(artist_ids)	Get up to 50 artists
top_tracks(artist_id, *, market)	Artist's top tracks
albums(artist_id, *, include_groups, market, limit, offset)	Artist's discography
related(artist_id)	Related artists

Albums — sp.albums

Method	Description
find(query, *, limit, offset, market)	Search for albums
get(album_id, *, market)	Get a single album
get_many(album_ids, *, market)	Get up to 20 albums
tracks(album_id, *, limit, offset, market)	Tracks in an album
new_releases(*, country, limit, offset)	New album releases

Playlists — sp.playlists

Method	Description
find(query, *, limit, offset)	Search for playlists
get(playlist_id, *, market)	Get a single playlist
list(*, user_id, limit, offset)	Current user's (or another user's) playlists
tracks(playlist_id, *, market, fields, limit, offset, additional_types)	Get playlist tracks
create(name, *, public, collaborative, description, user_id)	Create a playlist
update(playlist_id, *, name, public, collaborative, description)	Update playlist details
add(playlist_id, uris, *, position)	Add items to a playlist
replace(playlist_id, uris)	Replace all playlist items
remove(playlist_id, uris)	Remove items from a playlist
reorder(playlist_id, *, range_start, insert_before, range_length, snapshot_id)	Reorder items
cover_image(playlist_id)	Get playlist cover images

Player — sp.player

Method	Description
state(*, market)	Current playback state
play(*, device_id, context_uri, uris, offset, position_ms)	Start/resume playback
pause(*, device_id)	Pause playback
skip(*, device_id)	Skip to next track
previous(*, device_id)	Skip to previous track
seek(position_ms, *, device_id)	Seek to position
repeat(state, *, device_id)	Set repeat mode (track, context, off)
shuffle(state, *, device_id)	Toggle shuffle
volume(volume_percent, *, device_id)	Set volume (0–100)
queue()	Get the player queue
add_to_queue(uri, *, device_id)	Add a track/episode to the queue
transfer(device_id, *, play)	Transfer playback to another device
devices()	List available devices
recently_played(*, limit, after, before)	Recently played tracks

Library — sp.library

Method	Description
saved_tracks(*, limit, offset, market)	User's saved tracks
saved_albums(*, limit, offset, market)	User's saved albums
saved_shows(*, limit, offset)	User's saved shows
saved_episodes(*, limit, offset)	User's saved episodes
save_tracks(track_ids)	Save tracks
remove_tracks(track_ids)	Remove saved tracks
save_albums(album_ids)	Save albums
remove_albums(album_ids)	Remove saved albums
save_shows(show_ids)	Save shows
remove_shows(show_ids)	Remove saved shows
save_episodes(episode_ids)	Save episodes
remove_episodes(episode_ids)	Remove saved episodes
check_tracks(track_ids)	Check if tracks are saved
check_albums(album_ids)	Check if albums are saved
check_shows(show_ids)	Check if shows are saved
check_episodes(episode_ids)	Check if episodes are saved
top_tracks(*, time_range, limit, offset)	User's top tracks
top_artists(*, time_range, limit, offset)	User's top artists

time_range accepts "short_term", "medium_term", or "long_term".

Shows — sp.shows

Method	Description
find(query, *, limit, offset, market)	Search for shows
get(show_id, *, market)	Get a single show
get_many(show_ids, *, market)	Get multiple shows
episodes(show_id, *, market, limit, offset)	Episodes for a show

Episodes — sp.episodes

Method	Description
find(query, *, limit, offset, market)	Search for episodes
get(episode_id, *, market)	Get a single episode
get_many(episode_ids, *, market)	Get multiple episodes

Users — sp.users

Method	Description
me()	Current user's profile
get(user_id)	A public user's profile
following(*, type, limit, after)	Artists/users the current user follows
follow(type, ids)	Follow artists or users
unfollow(type, ids)	Unfollow artists or users
check_following(type, ids)	Check if following artists or users

Retries

Spotify API requests automatically retry rate limits (429) and temporary server errors (500, 502, 503, 504). Rate limits honor Spotify's Retry-After header. Server errors are only retried for idempotent HTTP methods to avoid duplicating mutations. Configure the defaults with max_retries and retry_backoff_seconds when constructing Spotifyify.

Use a request-context retry hook when planned retries should be reported to a caller before spotifyify sleeps and sends the request again. The hook is isolated per async task, so one shared Spotifyify instance can be used by concurrent conversations. retry_number is one-based and retry_at contains the planned retry time in UTC:

from spotifyify import RetryEvent

async def on_retry(event: RetryEvent) -> None:
    await sse_bus.emit(
        conversation_id,
        {
            "status_code": event.status_code,
            "retry_number": event.retry_number,
            "max_retries": event.max_retries,
            "retry_after": event.retry_after,
            "retry_at": event.retry_at.isoformat(),
        },
    )

async with spotify.session(on_retry=on_retry):
    track = await spotify.tracks.get(track_id)

If all retries are exhausted and Spotify still returns 429, spotifyify raises SpotifyRateLimitError. It subclasses SpotifyAPIError and exposes retry_after and retry_at from Spotify's final Retry-After response header:

from spotifyify import SpotifyRateLimitError

try:
    track = await spotify.tracks.get(track_id)
except SpotifyRateLimitError as exc:
    await sse_bus.emit(
        conversation_id,
        {
            "status_code": exc.status_code,
            "retry_after": exc.retry_after,
            "retry_at": exc.retry_at.isoformat() if exc.retry_at else None,
            "message": exc.message,
        },
    )

Request-scoped options can be combined without nested context managers:

async with spotify.session(access_token=end_user_access_token, on_retry=on_retry):
    me = await spotify.users.me()

Scopes

Use SpotifyScope to declare the OAuth scopes your app requires:

from spotifyify import SpotifyScope

SpotifyScope.USER_READ_PLAYBACK_STATE
SpotifyScope.USER_MODIFY_PLAYBACK_STATE
SpotifyScope.USER_LIBRARY_READ
SpotifyScope.USER_LIBRARY_MODIFY
SpotifyScope.USER_TOP_READ
SpotifyScope.USER_READ_RECENTLY_PLAYED
SpotifyScope.PLAYLIST_MODIFY_PUBLIC
SpotifyScope.PLAYLIST_MODIFY_PRIVATE
SpotifyScope.PLAYLIST_READ_PRIVATE

Scopes can also be passed as plain strings.

CLI

spotifyify ships with an optional Typer-based command line interface. It uses the same environment variables and .env loading as the Python client, so the configuration from above also applies to terminal usage.

Install it for the current project:

uv add "spotifyify[cli]"

Install it globally so spotifyify --help works from any directory:

uv tool install "spotifyify[cli]"
spotifyify --help

For local development from this checkout, install the current working tree as a global uv tool:

uv tool install --from . "spotifyify[cli]"
spotifyify --help

Or run the checkout directly without installing a global command:

uv sync --extra cli
uv run spotifyify --help

Output contract

Every command writes JSON to stdout and nothing else, regardless of whether stdout is a terminal — so piping through tee or capturing the output cannot change its shape.

Format	Always JSON
Shape	A JSON array of row objects whose keys are the command's declared columns, in a fixed order
Encoding	UTF-8, no ANSI escapes, no pager, no prompts
Errors	Plain text on stderr
Exit codes	0 ok, 1 API error, 2 usage error, 3 auth error, 4 no match

spotifyify tracks search "Ikkimel" --limit 2

[
  {
    "id": "4H0ly29pj5g6vMKum5kkhu",
    "name": "WHO'S THAT",
    "artists": ["Ikkimel"],
    "album.name": "WHO'S THAT",
    "uri": "spotify:track:4H0ly29pj5g6vMKum5kkhu"
  }
]

Set SPOTIFYIFY_RAW=1 to get the untouched Spotify payload instead, for debugging paging metadata or a field that is not a declared column.

SPOTIFYIFY_RAW=1 spotifyify tracks search "Daft Punk" --limit 1

PowerShell:

$env:SPOTIFYIFY_RAW = "1"
spotifyify tracks search "Daft Punk" --limit 1
Remove-Item Env:SPOTIFYIFY_RAW

Command discovery

Use the standard --help option at the root, group, or command level:

spotifyify --help
spotifyify artists --help
spotifyify artists get --help

The short form -h works at every level as well.

Resource groups use plural names consistently (artists, tracks, albums), and each command has one canonical spelling.

Everyday usage

The CLI mirrors the public namespace API from spotifyify.namespaces:

spotifyify tracks search "Daft Punk" --limit 5
spotifyify albums get 4aawyAB9vmqN3uQ7FjRGTy
spotifyify playlists list
spotifyify player state

To find something and play it without a separate lookup:

spotifyify play --artist Ikkimel --track "WHO'S THAT"

A track name (or free text) plays that one track; without one, --album plays the album and --artist alone plays the artist. If Spotify reports no active device, the CLI picks a controllable one and retries.

Mutations return the new state

Commands that change something report the state they produced, so no follow-up read is needed:

spotifyify player play --uri spotify:track:TRACK_ID

[{"state": "playing", "track": "HAMPELMANN", "artists": ["Ikkimel"], "device": "Wohnzimmer"}]

Playback commands briefly wait for Spotify to apply the change before reporting; pass --no-wait to skip that and read immediately. Library and follow mutations report the resulting saved/following state, and playlist mutations report the new snapshot and length.

Filtering

spotifyify tracks search "Daft Punk" --limit 3 --field id,name,uri
spotifyify playlists tracks PLAYLIST_ID --spotify-fields "items(track(id,name))"

Option	Effect
--field, -f	Replace the declared columns with the given field paths
--spotify-fields	Server-side filter applied by Spotify before it sends the response

Rows otherwise keep the order Spotify returned them in.

--field is a client-side output projection and can be repeated or receive a comma-separated list. Nested values use dotted paths:

spotifyify tracks get TRACK_ID --field id --field name --field album.name

Batching

Commands that take IDs or URIs are variadic and accept repeated or comma-separated values. One call fans out to as many API requests as Spotify's per-endpoint id limits require:

spotifyify tracks get ID_1 ID_2 ID_3
spotifyify albums get ID_1,ID_2
spotifyify playlists add PLAYLIST_ID spotify:track:ID_1 spotify:track:ID_2
spotifyify player add-to-queue spotify:track:ID_1 spotify:track:ID_2
spotifyify library save-tracks ID_1,ID_2,ID_3

Common options

Option	Description
--field, --fields, -f	Include only selected field paths
--limit, -l	Number of items to fetch, capped at Spotify's per-endpoint limits
--wait / --no-wait	Whether playback mutations wait for the change to take effect

Each command already requests the OAuth scopes it needs — there is no way to override that per call. When a command needs user authorization and no token is configured yet, the CLI uses the same interactive Authorization Code login and token cache as the Python client.

Global options

--market and --device-id apply to the whole invocation, so they go before the group name rather than on the individual command:

spotifyify --market DE tracks search "Daft Punk"
spotifyify --device-id kitchen player play --uri spotify:track:TRACK_ID

Option	Description	Env var fallback
--market, -m	ISO 3166-1 alpha-2 market code	SPOTIFYIFY_MARKET
--device-id	Target Spotify Connect device for playback commands	SPOTIFYIFY_DEVICE_ID

A flag always wins over its env var. Neither is required — omit both and Spotify falls back to its own default market and active device.

Command overview

Namespace	Commands
(top level)	play
tracks	search, get
artists	search, get, top-tracks, albums, related
albums	search, get, tracks, new-releases
playlists	search, get, list, tracks, create, update, add, replace, remove, reorder, cover-image
shows	search, get, episodes
episodes	search, get
library	saved-tracks, saved-albums, saved-shows, saved-episodes, top-tracks, top-artists, save-*, remove-*, check-* for tracks/albums/shows/episodes
player	state, play, pause, skip, previous, seek, repeat, shuffle, volume, queue, add-to-queue, transfer, devices, recently-played
users	me, get, following, follow, unfollow, check-following

spotifyify --help
spotifyify playlists create --help
spotifyify player play --help

Examples

See the examples/ directory for CLI recipes and runnable Python scripts:

• examples/cli/README.md — copy-paste CLI workflows for search, playback, playlists, library, batching, and JSON output
• examples/player/search_and_play.py — search for tracks and control playback with the Python API
• examples/playlists/manage_playlist.py — create and manage a playlist with the Python API
• examples/playlists/user_token_playlist.py — create playlists with caller-supplied user tokens
• examples/library/library_stats.py — explore your top tracks and saved library