MCP Veo2

🚀 Video Generation with Veo2's MCP

This project implements a Model Context Protocol (MCP) server that exposes Google's Veo2 video generation capabilities. It allows clients to generate videos from text prompts or images and access the generated videos via MCP resources.

✨ Features

• Video Generation: Supports generating high-quality videos from text prompts and images.
• Resource Access: Access generated videos via unique resource identifiers (e.g., videos://{id}) and provides example templates.
• Flexible Configuration: Offers various configuration options, including aspect ratio, person generation strategy, duration, etc.

📦 Installation

Project Dependencies

Make sure to install the following tools and libraries:

• Node.js (recommended version 14+)
• npm or yarn
• TypeScript (optional but recommended)

Clone the Repository

git clone https://github.com/your-repository.git
cd mcp-video-generation-with_veo2

Install Dependencies

npm install
# Or use yarn
yarn install

💻 Usage Examples

Initialize the Server

Run the following command to start the MCP server:

node src/index.ts

Call the Tool

Generate a Video

Generate from Text

{
  "prompt": "A magnificent city night view with twinkling lights",
  "config": {
    "aspectRatio": "16:9",
    "personGeneration": "dont_allow",
    "durationSeconds": 8
  }
}

Generate from an Image

{
  "image": "Base64-encoded image data...",
  "prompt": "Desert sunset"
}

List All Generated Videos

Access the resource videos:// to get a list of all videos.

📚 Documentation

Project Structure

• src/: Source code directory
  • index.ts: Program entry point
  • server.ts: MCP server configuration
  • config.ts: Configuration handling
  • tools/: MCP tool implementation
  • resources/: MCP resource implementation
  • services/: External service integration
  • utils/: Utility functions

Build the Project

npm run build

Development Mode

npm run dev

🔧 Technical Details

📄 License

MIT