Give any AI agent the power to make videos. Connect the VideoGen MCP server to Cursor, Claude, ChatGPT, or your own agent and generate videos, images, voiceovers, music, and avatars with natural language.
The full VideoGen API, exposed as tools your agent already knows how to use.
End-to-end video workflows (script to video, voiceover to video, slideshow to video), plus image, video clip, speech, music, sound effect, and avatar generation, project exports, remixing, and file management.
Cursor, Claude, ChatGPT, VS Code, Windsurf, Cline, Goose, and any other MCP client. One-click install for Cursor and VS Code, or paste one JSON block.
OAuth-capable hosts like ChatGPT and Claude connect with just the server URL: users sign in with VideoGen, no API key to paste. Review and revoke connected apps anytime.
Use the hosted server with nothing to install or update, or run the same tools locally over stdio when your agent needs access to files on your machine.
Video generation takes time. Composite tools start the operation and wait until it finishes before returning, so your agent gets a finished video, not a job id to babysit.
Every tool maps to a documented REST endpoint, so agents and code share one surface. Prefer a terminal? The VideoGen CLI drives the same operations.
Add the server to your MCP client config and start generating. OAuth hosts need only the URL.
{
"mcpServers": {
"videogen": {
"url": "https://mcp.videogen.io/mcp"
}
}
}Then ask your agent: "Generate a narrated video from this script with captions and background music."