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

# Captions MCP Server

> Learn how to install and use the Captions MCP Server

## What is an MCP Server?

MCP is an open protocol that standardizes how applications provide context to LLMs. Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect your devices to various peripherals and accessories, MCP provides a standardized way to connect AI models to different data sources and tools. [Learn more](https://modelcontextprotocol.io/introduction).

## What can I do with the Captions MCP Server?

Using [Claude](https://claude.com/) or [Cursor](https://www.cursor.com/), you can search Captions documentation.

This means you can enter a prompt like "What languages are supported for Dubbing in the Captions App?" Claude will search the Captions documentation for the answer.

## Get Started

To use the Captions MCP server you will need an [API key](https://captions.ai/help/api-reference/api) from your Captions account.

### Start the MCP server locally

<Steps>
  <Step title="Install the server through the CLI">
    Open Terminal and run the following command to install the server:

    ```
    npx @mintlify/mcp@latest add captions
    ```
  </Step>

  <Step title="Add the authentication token">
    You will be asked for your API key. Copy it from the [API Dashboard](https://desktop.captions.ai/api), paste it into the terminal, and hit `return` .

    ```
    What is the Captions x-api-key (API key)?
    ```
  </Step>

  <Step title="Add new MCP server to configs">
    A new MCP server will be created locally on your computer. Next, choose which programs you'd like to add.

    ```
    ?  Add new MCP server to configs: › Space to select (all selected by default). Return to submit. 
    ◉   Cursor
    ◉   Claude Desktop
    ```
  </Step>

  <Step title="Start your MCP server">
    Copy and paste the command (it will appear in the terminal, which you can copy and paste)

    ```
    node /Users/[username]/.mcp/captions/src/index.js
    ```
  </Step>
</Steps>

### Choose how you want to interact with the MCP

Next, open Claude or Cursor to start interacting with the Captions MCP Server.

**Instructions:**

* [Claude](https://modelcontextprotocol.io/quickstart/user)
* [Cursor](https://docs.cursor.com/context/model-context-protocol)