Lacuna Music

io.github.JOYLINK-LTDPublished by GitHub account JOYLINK-LTDv0.3.4Updated Oct 24, 2026

Generate AI music via the Lacuna Music API from MCP clients like Claude Desktop & Code.

The publisher's own server: Prediction Market Intelligence

VerifiedStreamable HTTPWeb executableMedia & DesignAI & ML

Overview

AI-generated overview

Lacuna Music MCP server that lets an assistant create AI music generation tasks, poll their status, and list available models.

What you can do
  • Create music generation tasks from a style description, title, optional lyrics or instrumental flag, model choice and tuning parameters
  • Check a generation task by id and read its status and, once ready, the finished tracks and audio URLs
  • Wait on a task by polling until it finishes or a timeout elapses
  • List the available music models with credit cost per task, tracks per task and language coverage
When to use it
Useful when an agent loop should produce original AI music or instrumentals — for example drafting a track from a style prompt and lyrics, then retrieving the audio links once generation completes.
Requirements
Node.js 18 or newer and the npm package run via npx over stdio, or the remote endpoint. A Lacuna API key in the LACUNA_API_KEY environment variable, from a Pro plan or above. LACUNA_BASE_URL optionally overrides the API base URL. Network access to the Lacuna API.
Before you install
The API key is a secret passed in the environment; keep it out of shared configs. Generation consumes credits, deducted when a task is created and refunded only if the upstream provider fails. Music access requires a paid Pro plan or above. Generated audio is hosted by Lacuna and returned as URLs.

Installation

In SourceWeft

  1. Open Lacuna Music in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/io.github.JOYLINK-LTD%2Flacuna-mcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "lacuna-music": {
      "type": "http",
      "url": "https://www.lacuna.fm/mcp"
    }
  }
}

README

lacuna-mcp

[npm version] [License: MIT] [Node.js]

Model Context Protocol server for the Lacuna Music API. Lets MCP-compatible clients (Claude Desktop, Claude Code, Cursor, Zed, Continue, etc.) generate AI music as part of an agent loop.

Looking for the SDK or CLI? See lacuna-sdk and lacuna-toolkit.

Installation

The server is published as a single npm package and runs over stdio. No install step is needed — your MCP client launches it via npx.

sh
npx lacuna-mcp

Configuration

The server reads its configuration from environment variables:

VariableRequiredDescription
LACUNA_API_KEYyesAPI key from your profile dashboard. Begins with lyr_live_.
LACUNA_BASE_URLnoOverride the API base URL (e.g. for staging).

Music API access requires the Pro plan or above.

Claude Desktop

Add an entry to claude_desktop_config.json:

json
{  "mcpServers": {    "lacuna": {      "command": "npx",      "args": ["-y", "lacuna-mcp"],      "env": {        "LACUNA_API_KEY": "lyr_live_..."      }    }  }}

The config file lives at:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

Claude Code

sh
claude mcp add lacuna -- npx -y lacuna-mcp

Then export the key in your shell or set it via your client's MCP settings.

Cursor / Zed / other MCP clients

Any client that speaks the Model Context Protocol over stdio can run npx -y lacuna-mcp as the command and pass LACUNA_API_KEY as an env var. Refer to your client's MCP setup docs for the exact JSON shape.

Tools

ToolDescription
generate_musicCreate a music generation task. Returns a pending task immediately.
get_generationRetrieve the current state of a generation task by id.
wait_for_generationPoll a task until it reaches a terminal state (ready or failed) or the timeout elapses.

generate_music

ParameterTypeDescription
stylestring (required)Style description, e.g. "pop, female vocal, 120 bpm, energetic".
titlestring (required)Track title.
lyricsstringRequired unless instrumental is true.
instrumentalbooleanGenerate an instrumental track. Default false.
model"aether" | "echo"Generation model. Defaults to aether (Lacuna Aether).
vocal_gender"m" | "f"Lead vocal hint (aether only).
negative_tagsstringStyle tags to avoid (aether only).
style_weightnumber (0–1)aether only.
weirdness_constraintnumber (0–1)aether only.
audio_weightnumber (0–1)aether only.

Credits are deducted on creation and refunded automatically if the upstream provider fails.

get_generation

ParameterTypeDescription
idstringGeneration task id, returned by generate_music.

wait_for_generation

ParameterTypeDescription
idstringGeneration task id.
poll_interval_secondsnumberPolling interval. Default 5.
timeout_secondsnumberTotal timeout. Default 600 (10 minutes).

Requirements

  • Node.js 18 or newer.
  • An API key from a Lacuna Pro plan or above.

License

MIT © Louis Tsang

Source: packages/mcp/README.md at commit 0b19ff9

Tools

3

generate_music

Generate music

Create an AI music generation task on Lacuna. Returns immediately with a `pending` task; poll `get_generation` to retrieve the finished tracks and their audio URLs.

Parametersmodelstylerequiredtitlerequiredlyricsaudio_weightinstrumentalstyle_weightvocal_gendernegative_tagsweirdness_constraint

get_generation

Get generation

Retrieve the current state of a music generation task by id. Status is `pending`, `ready`, or `failed`; audio URLs appear on `ready`.

Parametersidrequired

list_models

List models

List the music models available through the API, with their credit cost per task, tracks produced per task, and language coverage.

Takes no parameters.

Version history

1
  1. v0.3.4LatestSep 16, 2026

Score

100% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

Other Lacuna integrations

Prediction Market Intelligence

markets.lacunaOfficial publisher

A read-only remote MCP server that lets an assistant search Polymarket markets and inspect prices, order books, trader flow, movers, leaderboards and wallet P&L

Streamable HTTPWeb executable
Finance7 toolsOct 24, 2026

Lacuna Research MCP

Published by GitHub account tiptreesystems

Search Lacuna's ML research graph for papers, directions, authors, venues, and hypotheses.

Streamable HTTPWeb executable
Web Search & Scraping13 toolsNov 3, 2026