> ## Documentation Index
> Fetch the complete documentation index at: https://support.vocuno.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Use Vocuno in ChatGPT, Claude and more

> Connect Vocuno to ChatGPT, Claude, Cursor, Claude Code, Codex or the Vocuno CLI and create songs, covers, stems and audio edits from your AI assistant.

<Card title="Connect your AI tool" icon="plug" href="https://vocuno.com/mcp">
  Open vocuno.com/mcp
</Card>

You can use Vocuno from the AI assistant you already work in. Connect it once, sign in with your Vocuno account, and ask for a song, a cover, stems, a voice conversion, or an audio edit in plain language. Your assistant runs the Vocuno tool and gives you the result with a share link.

Vocuno works with ChatGPT, Claude (web, desktop, mobile, and Claude Code), Cursor, Codex, Perplexity, OpenClaw, Hermes Agent, and any other AI tool that supports remote MCP servers with OAuth sign-in. MCP (Model Context Protocol) is the open standard that lets AI assistants use outside tools.

## What you need to know

* **No API key.** You sign in with your Vocuno account, and you can revoke access at any time.
* **Same account, same credits.** Creations use the same credits as the Vocuno app, at the same prices. See [Credit Costs](/credit-costs).
* **Everything lands in your library.** Every song you make from an assistant is saved to your Vocuno library like any other creation, with a permanent vocuno.com share link and player.
* **100 free credits.** The first time you connect any AI tool to Vocuno, we add 100 credits to your account, once per account.

## Set it up

Go to [vocuno.com/mcp](https://vocuno.com/mcp) and pick your AI tool. The page shows the exact steps and buttons for each one.

<Tabs>
  <Tab title="ChatGPT">
    1. Add the **Vocuno** plugin from the ChatGPT plugin directory, or click **Install Vocuno plugin** on [vocuno.com/mcp](https://vocuno.com/mcp).
    2. Click **Add** and sign in with your Vocuno account.
    3. Ask ChatGPT to make a song or edit your audio.

    No connector URL or developer mode is needed.
  </Tab>

  <Tab title="Claude">
    1. On [vocuno.com/mcp](https://vocuno.com/mcp), choose Claude and click **Add Vocuno to Claude**. Claude opens with the Vocuno connector already filled in.
    2. Click **Add**, then **Connect**, and sign in with your Vocuno account.
    3. Ask Claude to make a song or edit your audio. Once connected, it works in Claude on the web, desktop, and mobile.

    To add it by hand, use this connector URL: `https://vocuno.com/mcp`
  </Tab>

  <Tab title="Cursor">
    1. On [vocuno.com/mcp](https://vocuno.com/mcp), choose Cursor and click **Add to Cursor**. You can also paste this into `~/.cursor/mcp.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "vocuno": {
          "url": "https://vocuno.com/mcp"
        }
      }
    }
    ```

    2. Open **Cursor Settings → Tools & MCP**, find Vocuno, and click **Connect** to sign in.
    3. In Agent mode, ask Cursor for a song. It returns a share link.
  </Tab>

  <Tab title="Claude Code, Codex & the CLI">
    Terminal agents such as Claude Code and Codex work best with the Vocuno CLI. It needs Node.js 20 or later.

    ```bash theme={null}
    npm i -g vocuno
    vocuno auth login
    vocuno generate "an upbeat synthwave track" --wait
    ```

    `vocuno auth login` opens your browser to sign in, with no API key to create. The `--wait` flag follows progress until the track is ready. Then `vocuno download` grabs the audio, `vocuno credits` shows your balance, and `vocuno tools` lists everything else.

    To teach your coding agent how to use Vocuno, add the companion skills:

    ```bash theme={null}
    npx skills add vocuno/skills
    ```

    You can also copy the ready-made setup message from [vocuno.com/mcp](https://vocuno.com/mcp) and paste it to your agent. It installs the CLI, signs you in, and adds the skills for you.
  </Tab>

  <Tab title="Other AI tools">
    Add Vocuno to your tool's MCP settings with the Streamable HTTP transport and this URL:

    ```
    https://vocuno.com/mcp
    ```

    Your tool opens a Vocuno sign-in page. Approve the connection, and you're linked. [vocuno.com/mcp](https://vocuno.com/mcp) has step-by-step guides for Perplexity, OpenClaw, and Hermes Agent.
  </Tab>
</Tabs>

## What your assistant can do

* Generate full songs with Suno, Mureka, MiniMax, Lyria, Udio, and more.
* **Remix any song** in one step: name the song, paste a link, or hand over your own audio file. Vocuno fetches the audio and starts the remix, with your own lyrics, a new style, or as an instrumental.
* Separate stems and remove reverb from any track.
* Convert voices and clone a voice from a sample.
* Trim, fade, mix, master, and convert audio files.
* Extend songs, replace sections, and detect chords and BPM.

In compatible AI tools, your songs show as interactive music cards: cover art, a player, a switch between versions, and any stems already separated, with a link back to your library.

<Note>
  Remixes and covers made with the Vocuno engine cost 150 credits and need a plan that includes Cover / Remix, the same as on the website. See [Billing & Subscriptions](/billing-and-subscriptions#what-each-plan-includes).
</Note>

## Credits and limits

* If a request needs more credits or a paid plan, your assistant gives you a link to the pricing page so you can top up or upgrade and carry on.
* On the Studio plan, a song that has to wait before it starts tells your assistant when it can begin. Nothing is charged for the wait.
* Most songs finish in 1 to 3 minutes. Your assistant shares a preview link as soon as one exists, and it upgrades to the final track automatically.

## Disconnect an AI tool

Go to [vocuno.com/settings](https://vocuno.com/settings) and open **Connected apps**. Every app you connected is listed with when you connected it and when it was last used. Click **Revoke** next to an app. It loses access to your account within a minute and has to be connected again from scratch.

## Frequently asked questions

<AccordionGroup>
  <Accordion title="Is this the same as the Vocuno Agent?">
    No. [Vai, the Vocuno Agent](/agent/overview), is the AI chat built into vocuno.com and the mobile app. This page is about using Vocuno from an assistant you already use, such as ChatGPT or Claude. Both use the same account, library, and credits.
  </Accordion>

  <Accordion title="Do I need a paid plan to connect?">
    No. Any Vocuno account can connect. What you can create depends on your plan and credits, exactly as on the website.
  </Accordion>

  <Accordion title="The assistant says it can't read my audio file. What should I do?">
    The assistant tells you in plain words what went wrong, for example a file that is not audio or an upload that never finished. Try uploading the file again, or use an MP3 or WAV.
  </Accordion>
</AccordionGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.