Skip to main content
Glama
ReccoBeats MCP Server

ReccoBeats MCP Server

by FDeRubeis

ReccoBeats MCP Server

An MCP server that provides integration with ReccoBeats, a free music database and recommendation API. This server lets the AI assistant in any MCP client (such as Claude Desktop, Claude Code, Cursor or VS Code) look up music, read how tracks sound and recommend new ones. It doesn't need an API key or an account.

Unofficial project. This server is not affiliated with or endorsed by ReccoBeats.

What you can do

Ask your assistant things like:

  • "Find Bohemian Rhapsody by Pentatonix and tell me its tempo and key."

  • "Recommend 10 tracks like this one, but more upbeat: https://open.spotify.com/track/0lrkMvRttmoXjMNS8YONvj"

  • "List Daft Punk's albums and compare the energy of their tracks."

  • "Analyse ~/Music/demo.mp3 and suggest released songs that sound similar."

In plain terms, the server enables the assistant to:

  • Search for tracks, artists and albums by name.

  • Look up details, such as a track's artists, duration and popularity, or an album's label and release date.

  • Read audio features: numbers that describe how a track sounds, such as energy, mood, danceability, tempo and key.

  • Get recommendations from up to 5 seed tracks, optionally steered toward a target sound (for example "calmer" or "more danceable").

  • Analyse your own audio files by uploading a short local clip to ReccoBeats and getting its audio features back.

When talking to the assistant, you can refer to tracks, albums and artists by their name. To avoid ambiguity, you can also use ReccoBeats IDs or Spotify IDs, URIs (spotify:track:…) or links (https://open.spotify.com/…). For tracks you can also use ISRCs.

Related MCP server: spotify-mcp

Tools

Tool

What it does

search_tracks

Search tracks by title, optionally filtered by artist.

search_artists

Search artists by name.

search_albums

Search albums by name.

get_tracks

Get details for up to 40 tracks (by ReccoBeats ID, Spotify ID or ISRC).

get_artists

Get details for up to 40 artists.

get_albums

Get details for up to 40 albums.

get_audio_features

Get audio features (energy, valence, tempo, key, …) for up to 40 tracks.

get_artist_tracks

List an artist's tracks.

get_artist_albums

List an artist's albums, singles and compilations.

get_album_tracks

List the tracks on an album.

get_track_albums

List the albums a track appears on.

get_recommendations

Recommend tracks similar to 1–5 seed tracks, optionally steered by target audio features.

extract_audio_features

Upload a local audio file and return its audio features.

extract_audio_features accepts MP3, OGG (Vorbis), WAV and AIFF/AIFC files up to 5 MB, and the server checks the format from the file's content before uploading. ReccoBeats analyses at most 30 seconds of audio.

Usage

Requirements

  • Python 3.12 or later

  • uv

Install

git clone <repository-url> reccobeats-mcp
cd reccobeats-mcp
uv sync

Connect your client

Add the following configuration to your MCP client (e.g. Claude Desktop, Claude Code, Cursor or VS Code).

{
  "mcpServers": {
    "reccobeats": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/reccobeats-mcp", "reccobeats-mcp"]
    }
  }
}

Restart the client, then ask it something like "Search ReccoBeats for Bohemian Rhapsody" to check that it works.

Run as an HTTP server

By default the client starts the server itself and talks to it over stdio. You can instead run one long-lived server over Streamable HTTP, for example for a remote client or for several clients sharing one process:

uv run reccobeats-mcp --transport streamable-http
# ReccoBeats MCP server listening on http://127.0.0.1:8000/mcp

Then point your client at http://127.0.0.1:8000/mcp.

Security: the HTTP endpoint has no authentication. It listens only on your own machine by default. If you bind another address (such as --host 0.0.0.0) or expose it through a tunnel, put authentication and rate limiting in front of it. In HTTP mode, extract_audio_features reads files from the server's disk, not the caller's.

Configuration

Options can be set with a command-line flag or an environment variable. If both are set, the flag wins.

Flag

Environment variable

Default

Description

--transport

RECCOBEATS_TRANSPORT

stdio

stdio or streamable-http.

--host

RECCOBEATS_HOST

127.0.0.1

Address to listen on in HTTP mode.

--port

RECCOBEATS_PORT

8000

Port to listen on in HTTP mode.

--path

RECCOBEATS_PATH

/mcp

URL path of the MCP endpoint in HTTP mode.

RECCOBEATS_BASE_URL

https://api.reccobeats.com/v1

ReccoBeats API base URL.

RECCOBEATS_TIMEOUT

30

Timeout for ReccoBeats API requests, in seconds.

To set environment variables from a client configuration, add an env block next to args:

"env": { "RECCOBEATS_TIMEOUT": "60" }

License and disclaimer

Released under the MIT License.

This is an independent, community project. It is not affiliated with, endorsed by or sponsored by ReccoBeats. "ReccoBeats" and "Spotify" are names of their respective owners and are used here only to describe what the software works with. Use of the ReccoBeats API is subject to ReccoBeats' own terms.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      D
      maintenance
      Enables interaction with Spotify through MCP, providing tools for searching tracks, artists, albums, playlists, and accessing user data like top tracks and recently played.
      -
    • A
      license
      C
      quality
      C
      maintenance
      Enables MCP clients to control playback and queues, search and discover tracks, manage saved items and followed artists, and perform bulk playlist edits, including dry-run previews for destructive operations.
      36
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables MCP clients to search Spotify's track catalog and retrieve the connected user's recently played tracks through natural language tool calls.
      127 npm
      MIT