Vidofy — Claude Code MCP server: media generation and setup
Vidofy gives AI coding agents access to models that create video, images, audio and speech. Use it to generate media from prompts or work with existing files.
claude mcp add --transport http mcp https://vidofy.ai/mcp-appWhat Vidofy does
The Vidofy MCP server connects an agent to tools for video and image generation, photo editing, video effects, lip sync, speech synthesis and voice cloning. Agents can inspect available modes, compare models, check input requirements, estimate a job's price and retrieve the finished media. Other tools expose account balance and recent generation costs.
A typical workflow starts by selecting a mode and model, then checking its input schema and estimating the price before submitting a job through generate. This call deducts coins at submission, including when a generation later fails. The agent checks progress separately, and generated media stays private unless a public link is explicitly requested.
Connect over HTTP using the service's remote endpoint, or run the local server with npx -y @vidofy/mcp and a personal VIDOFY_TOKEN. Jobs draw from the user's own Vidofy coin balance at the service's listed rates. Startup rejects VIDOFY_API_KEY; the server's tools also cannot purchase coins or change a subscription.
What it helps with
- Generate a video from a text prompt
- Create an image or edit an existing photo
- Turn text into speech or clone a voice
- Estimate generation costs before submitting a job
- Track a generation and retrieve its output
Example prompts for the agent
- Find a video generation mode and estimate the cost of a clip featuring a red bicycle.
- Choose a model for editing this image and inspect its file requirements.
- Check the status of my generation and retrieve the finished result.
How to install Vidofy
1Platform CLI: remote server
claude mcp add --transport http mcp https://vidofy.ai/mcp-app2Platform CLI
claude mcp add --env VIDOFY_TOKEN=<VIDOFY_TOKEN> --transport stdio vidofy -- npx -y @vidofy/mcpReplace <…> with your own values
1Platform CLI: remote server
codex mcp add vidofy --url https://vidofy.ai/mcp-app2Platform CLI
codex mcp add vidofy --env VIDOFY_TOKEN=<VIDOFY_TOKEN> -- npx -y @vidofy/mcpReplace <…> with your own values
Config file
{
"mcpServers": {
"vidofy": {
"url": "https://vidofy.ai/mcp-app"
}
}
}{
"mcpServers": {
"vidofy": {
"command": "npx",
"args": [
"-y",
"@vidofy/mcp"
],
"env": {
"VIDOFY_TOKEN": "<VIDOFY_TOKEN>"
}
}
}
}Replace <…> with your own values
Config file
{
"servers": {
"vidofy": {
"type": "http",
"url": "https://vidofy.ai/mcp-app"
}
}
}{
"servers": {
"vidofy": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@vidofy/mcp"
],
"env": {
"VIDOFY_TOKEN": "<VIDOFY_TOKEN>"
}
}
}
}Replace <…> with your own values
1Platform CLI: remote server
gemini mcp add -t http vidofy https://vidofy.ai/mcp-app2Platform CLI
gemini mcp add -e VIDOFY_TOKEN=<VIDOFY_TOKEN> vidofy npx -y @vidofy/mcpReplace <…> with your own values
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about Vidofy
How do I install Vidofy in Claude Code?
For remote access, run claude mcp add --transport http vidofy https://vidofy.ai/mcp-app. To run the server locally with your personal token, use claude mcp add --env VIDOFY_TOKEN=<VIDOFY_TOKEN> --transport stdio vidofy -- npx -y @vidofy/mcp instead.
What does the Vidofy MCP server do?
It exposes media generation tools to AI agents, covering video, images, audio and speech. Agents can inspect models, calculate costs, submit jobs and retrieve their outputs.
When are Vidofy coins charged?
Coins are deducted when generate submits a job, not when the output succeeds. Use estimate_cost before submission to check the expected charge.
