search_songLABSCO SUMMARY
The story behind a song, with its sourcing attached.
Genius is two things at once: an official API and a large body of community writing. Annotations arrive tagged with a trust level, so an agent can tell an artist-verified explanation from an anonymous guess instead of flattening both into 'the internet says'.
What it is
An MCP server over Genius.com, using the official Genius API with your access token and the lyricsgenius Python library for endpoints the official API does not cover.
What you get
- Songs and artists searched by name
- Full song metadata — title, album, release date, lyrics state and Genius's editorial description
- Artist profiles with bio, follower count and verification status
- Line-level annotations explaining specific lyric fragments, each carrying a trust level so a model knows how much weight to give it
- A single annotation read in detail
- Song relationships — what a track samples, interpolates, covers or remixes, and what later sampled it
- Credits: writers, producers, featured artists and custom roles such as mixing engineer, studio and label
- The questions and answers attached to a song
Requirements
A Genius API access token. Runs with Python or from the included Docker image, in either transport mode. Worth knowing which half you are on: the tools built on lyricsgenius reach Genius's undocumented public API, which is not part of the official contract and can change without notice — those are the results that will break first.
Setup effort
Build from source — git clone https://github.com/federicogarciav/genius-mcp.git
USE CASES
01
Finding the right version of a song
Search by name, then read the full metadata to confirm album and release date.
02
Understanding a line that does not make sense
Pull the song's annotations and open the one covering that fragment.
03
Tracing where a track came from
Read its relationships — what it samples, interpolates or covers, and who has sampled it since.
04
Finding out who actually made it
Read the credits: writers, producers, features and the studio roles.
05
Researching an artist properly
Search the artist, read the profile, then work through the songs attached to it.
06
Answering the questions fans already asked
Read the song's questions and answers before writing your own explanation.
14 tools
Listed in project docs. Your client may show a different list.
get_song_detailsget_song_annotationsget_annotation_detailget_song_questions_and_answersget_song_relationshipsget_song_creditssearch_artistget_artist_detailsget_artist_songsget_artist_albumssearch_albumget_album_detailsget_cover_art_annotationsBEFORE YOU INSTALL
What to know first
Tools you get
14 tools named in the project's documentation.
Requires
GENIUS_ACCESS_TOKEN in your environment
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
TERMINAL
git clone https://github.com/federicogarciav/genius-mcp.git
cd genius-mcp
pip install -r requirements.txtMCP Clients
Claude Code
This route needs the server running in Streamable HTTP mode: with STREAMABLE_HTTP=true in your .env (copy it from .env.example and add your token), run python main.py in the clone — it listens on http://127.0.0.1:8080. Then register it:
CLAUDE CODE · TERMINAL
claude mcp add genius --transport http http://127.0.0.1:8080/mcpClaude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"genius": {
"command": "python",
"args": [
"<YOUR_PATH_TO_GENIUS_MCP>/main.py"
],
"env": {
"GENIUS_ACCESS_TOKEN": "<YOUR_GENIUS_ACCESS_TOKEN>",
"STREAMABLE_HTTP": "false"
}
}
}
}
