System Components
The main components of this system:
1. UVX MCP Server:
- Python-based cloud server
- Communicates directly with the Vidu API
- Provides complete video generation functionality
Installation and Configuration
Prerequisites
- Python 3.10 or newer
- UV/UVX
- Vidu API key: obtained from Vidu Platform (API Credits must be purchased from Vidu Platform)
Getting Dependencies
- Python
- Download and install from the official Python website
- Make sure Python is added to the system PATH
- UV/UVX
- Download UV / UVX using the methods below
- mac/linux
- Windows
How to Use the MCP Server
Get a Vidu API Key
- Visit Vidu Platform
- Register/log in to your account
- Create and copy your API key in your personal settings
Download the Required Dependencies
- Python: install Python 3.10 or higher
- UV/UVX: install the latest stable version of UV & UVX
Configure the MCP Client
- Open your MCP client (e.g., Claude for Desktop or Cursor)
- Locate the client settings
- Find and open the mcp_config.json file (or the corresponding configuration file)
- Add the configuration according to your chosen method:
- Save the configuration file
Restart the MCP Client or Refresh the MCP Server
- Fully close and reopen your MCP client
- Alternatively, if the client supports it, use the refresh MCP server option
Client-Specific Configuration
Claude for Desktop Configuration
- Open the Claude app
- Go to Claude > Settings > Developer > Edit Config
- Open the claude_desktop_config.json file
- Add the configuration above and save
- Restart the Claude app
- If the connection succeeds, the home page will not show any error, and the MCP setting indicator will turn green
- If the connection fails, the home page will show a connection failure message
Cursor Configuration
- Open the Cursor app
- Go to Settings > Model Context Protocol
- Add a new server
- Fill in the server details (same information as in the JSON configuration above)
- Save and restart, or refresh the MCP server
Usage Examples
Text-to-Video
Through Claude or Cursor, you can use natural language to describe the video you want to generate: Basic example:
Advanced example (with parameters):