orcadub-mcp-server — Claude Code MCP server: setup and usage
MCP server that automates video dubbing into other languages through OrcaRouter. It uploads footage, triggers the orca/dub model, polls progress, and downloads the final MP4.
claude mcp add orcadub-mcp-server -e ORCADUB_API_KEY=<ORCADUB_API_KEY> -- npx -y @orcadub/cliWhat orcadub-mcp-server does
The server gives AI agents access to the orca/dub model via the OrcaRouter gateway. The workflow runs end to end. The agent takes a local clip or a video link, submits a translation job that keeps the speaker's original voice, monitors processing, and downloads the resulting MP4. The CLI package @orcadub/cli runs directly as an stdio server or inside a Docker container.
Each dubbing job accepts source and target languages alongside extra parameters. You can retain background music with preserve_bgm, disable watermarks, set 1080p resolution, or inject custom glossary terms. Local filesystem access is required by dub_upload and dub_download. When running through Docker, you must mount the working directory using the -v option so the container sees your files.
Without an ORCADUB_API_KEY, tool calls will not process video. The server still boots up, yet every call redirects to the OrcaRouter console registration page. Billing depends on runtime: each minute of source video incurs charges.
What it helps with
- translate spoken video audio while preserving the original voice
- keep background music intact during dubbing via
preserve_bgm - automate video uploading, job polling, and MP4 downloading inside agent workflows
- connect the
orca/dubpipeline to coding assistants over stdio or Docker
Example prompts for the agent
- Dub
./clip.mp4from English to Japanese withorca/duband keep the background music. - Upload
./clip.mp4, start dubbing into Chinese, and save the result to./out.mp4. - Check the status of dubbing job
<id>and download the finished video to./out.mp4.
How to install orcadub-mcp-server
Platform CLI
claude mcp add orcadub-mcp-server -e ORCADUB_API_KEY=<ORCADUB_API_KEY> -- npx -y @orcadub/cliReplace <…> with your own values
Platform CLI
codex mcp add orcadub-mcp-server --env ORCADUB_API_KEY=<ORCADUB_API_KEY> -- npx -y @orcadub/cliReplace <…> with your own values
Config file
{
"mcpServers": {
"orcadub-mcp-server": {
"command": "npx",
"args": [
"-y",
"@orcadub/cli"
],
"env": {
"ORCADUB_API_KEY": "<ORCADUB_API_KEY>"
}
}
}
}Replace <…> with your own values
Config file
{
"servers": {
"orcadub-mcp-server": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@orcadub/cli"
],
"env": {
"ORCADUB_API_KEY": "<ORCADUB_API_KEY>"
}
}
}
}Replace <…> with your own values
Platform CLI
gemini mcp add -e ORCADUB_API_KEY=<ORCADUB_API_KEY> orcadub-mcp-server npx -y @orcadub/cliReplace <…> 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 orcadub-mcp-server
How do I install orcadub-mcp-server in Claude Code?
Run claude mcp add orcadub-mcp-server -e ORCADUB_API_KEY=<ORCADUB_API_KEY> -- npx -y @orcadub/cli in your terminal. Replace <ORCADUB_API_KEY> with your key from the OrcaRouter console.
What does orcadub-mcp-server do?
It gives AI assistants tools to dub video content using the orca/dub model. The server uploads the source media, submits a dubbing job that preserves speaker voice characteristics, checks the status, and saves the final MP4.
What is required to run the server and how is it billed?
You need an ORCADUB_API_KEY from the OrcaRouter console at https://www.orcarouter.ai/console. The server starts without a key, but every tool call redirects to sign-up. Usage is billed per minute of input footage.