generate_lyrics
Generate lyrics with a given prompt then return the title and lyrics text to the client directly.
LABSCO SUMMARY
The timeout is a real setting, not boilerplate.
Songs take long enough that the default 60-second window is the first thing most people hit, which is why the vendor's own sample config sets 300. Plan for a request that runs for minutes and returns one artefact, rather than a chat-speed round trip — and remember each attempt spends credits.
What it is
Mureka's official MCP server over its lyrics, song and instrumental generation APIs. Four tools show up in the client once it loads, and the client plays or saves what comes back.
What you get
- Lyrics written to a prompt — a theme, an occasion, a person
- A song generated from lyrics you just produced, or from a plain description
- Background music and instrumentals described by mood and setting rather than by genre tags
- A generation timeout you control with
TIME_OUT_SECONDS, which defaults to 60 seconds; the sample configuration raises it to 300 - Client-side logs at a documented path on Windows and macOS when a generation does not come back
Requirements
A Mureka API key from platform.mureka.ai in MUREKA_API_KEY, with MUREKA_API_URL set to https://api.mureka.ai. Run it with uvx mureka-mcp — the package is mureka-mcp, so uv is the only install step. Generation consumes Mureka credits, which the README flags before the first example.
Setup effort
One command plus a key — uvx mureka-mcp, then supply credentials
USE CASES
01
Making a song for an occasion
Describe the person and the moment, get lyrics, then turn those same lyrics into a finished track without leaving the conversation.
02
Scoring a room or a video
Ask for instrumental background music by setting and mood — the worked example is music suitable for a coffee shop.
03
Driving it from your own agent
Connect it through the OpenAI Agents SDK with the same key and watch the composition steps in the console.
4 tools
Listed in project docs. Your client may show a different list.
generate_song
Generate song based on the lyrics text and save the output audio file to a given directory.
generate_instrumental
Generate background music(instrumental) based on the prompt text and save the output audio file to a given directory.
play_audio
Play an audio file. Supports WAV and MP3 formats.
BEFORE YOU INSTALL
What to know first
Tools you get
4 tools named in the project's documentation.
Requires
- uvx on PATH
- MUREKA_API_KEY in your environment
INSTALLATION
Connect your client
Server setup
TERMINAL
uvx mureka-mcpMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"Mureka": {
"command": "uvx",
"args": [
"mureka-mcp"
],
"env": {
"MUREKA_API_KEY": "<YOUR_MUREKA_API_KEY>",
"MUREKA_API_URL": "https://api.mureka.ai",
"TIME_OUT_SECONDS": "300"
}
}
}
}OpenAI Agents SDK
OPENAI AGENTS SDK · TERMINAL
git clone https://github.com/SkyworkAI/Mureka-mcp.git
cd Mureka-mcp
pip install openai-agentsIn openapi_agents_example/main.py, set MUREKA_API_KEY to your Mureka API key
OPENAI AGENTS SDK · TERMINAL
export OPENAI_API_KEY=<YOUR_OPENAI_API_KEY>
python openapi_agents_example/main.py
