Suno MCP

What suno-mcp does

suno-mcp provides MCP access to Suno operations through RunAPI. Its tools cover text-to-music, text-to-sound, lyric generation, voice generation, audio conversion, covers, remastering, stem separation, mashups, stitching, visualization, personas, and other Suno workflows.

The package is scoped to the Suno model family instead of exposing the broader RunAPI catalog. Nine model variants are available across the supported endpoints, although model choices vary by tool. Some operations accept models such as suno-v4, suno-v4.5, suno-v5, suno-v5.5, suno-v6, and related variants.

How it works

Install the package as an MCP server over stdio, then configure an MCP host to launch it with npx. An assistant can call check_pricing before submitting work, create a task with one of the asynchronous operation tools, and either wait for a terminal result or receive a task identifier immediately.

Asynchronous tools return task information and, depending on the operation, output URLs or a result payload. Use get_task to retrieve the latest status and payload for a task that was submitted without waiting. generate_persona, boost_style, get_timestamped_lyrics, and generate_voice run synchronously and return their operation results directly.

The server uses RunAPI authentication. Interactive login opens a browser and stores credentials in ~/.config/runapi/config.json; the same configuration can be used by the RunAPI CLI. For headless or CI environments, set RUNAPI_API_KEY before starting the MCP host. Pricing lookup remains available without authentication.

Setup and configuration

For Claude Code, add the server with:

claude mcp add suno -s user -- npx -y @runapi.ai/suno-mcp

Use project scope instead when the configuration should be shared with a repository. Other MCP hosts can launch the package with the equivalent stdio configuration:

{
  "mcpServers": {
    "suno": {
      "command": "npx",
      "args": ["-y", "@runapi.ai/suno-mcp"]
    }
  }
}

After installation, ask the assistant to call the login tool for interactive authentication. A pre-provisioned configuration file may contain an apiKey field, but real credentials should not be committed to source control.

Tools and capabilities

The operation tools include:

• text_to_music and text_to_sound for generating music or sound from text.
• generate_lyrics, get_timestamped_lyrics, generate_voice, and generate_persona for lyric, voice, timing, and persona workflows.
• convert_audio, cover_audio, remaster_audio, separate_audio_stems, stitch_audio, and add_samples for audio processing.
• blend_lyrics, extend_music, inspire_music, create_mashup, visualize_music, and boost_style for additional Suno transformations.
• get_task for checking an existing task and check_pricing for pricing by model and endpoint.

Most creation and processing tools can accept a wait option so the caller can choose between polling until completion and returning after submission. Tools that expose output URLs return them with the task status; lyric and blend operations return their result payloads.

Limitations and notes

Authentication is needed for task creation and status polling, while check_pricing is the unauthenticated exception. Model availability and the set of models accepted by each endpoint can change between package releases, so pricing lookup or the Suno model page should be used for the current catalog. The server communicates with Suno through RunAPI and does not provide the full RunAPI model collection.