remotion-media

Remotion Media MCP

An MCP (Model Context Protocol) server for AI-powered media generation. Generate images, videos, music, sound effects, speech, and subtitles directly from Claude or any MCP-compatible client.

All generated files save to public/ — ready for Remotion's staticFile() or any other use.

Quick Start

Install globally

npm install -g remotion-media-mcp

Add to Claude Code

If installed globally or via npx:

claude mcp add remotion-media \
  -e KIE_API_KEY=your-api-key \
  -- npx remotion-media-mcp

If cloned and built locally:

claude mcp add remotion-media \
  -e KIE_API_KEY=your-api-key \
  -- node /path/to/remotion-media-mcp/dist/index.js

With optional Airtable asset tracking:

claude mcp add remotion-media \
  -e KIE_API_KEY=your-api-key \
  -e AIRTABLE_API_KEY=your-airtable-pat \
  -e AIRTABLE_BASE_ID=appXXX \
  -e AIRTABLE_TABLE_NAME=Assets \
  -- npx remotion-media-mcp

By default this adds at user scope (-s user). Add -s project to scope it to the current project only.

Add to Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):

{
  "mcpServers": {
    "remotion-media": {
      "command": "npx",
      "args": ["remotion-media-mcp"],
      "env": {
        "KIE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Add to .mcp.json

{
  "mcpServers": {
    "remotion-media": {
      "command": "npx",
      "args": ["remotion-media-mcp"],
      "env": {
        "KIE_API_KEY": "your-api-key-here"
      }
    }
  }
}

API Key

Get a KIE_API_KEY from kie.ai — sign up and grab your key from the dashboard.

Tools

generate_image

Generate AI images via Nano Banana Pro.

generate_video_from_text

Text-to-video via Veo 3.1. Creates ~8 second clips.

generate_video_from_image

Animate a still image or transition between two images via Veo 3.1.

generate_music

Generate AI music via Suno.

generate_sound_effect

Generate sound effects via ElevenLabs SFX V2.

generate_speech

Text-to-speech via ElevenLabs TTS. 21 voices available.

generate_subtitles

Transcribe audio/video to SRT subtitles using local Whisper.

Requires whisper.cpp or OpenAI Whisper:

brew install whisper-cpp    # recommended
# or
pip install openai-whisper

list_assets

Browse assets in the Airtable asset library.

backup_asset

Back up a local file to the asset library. Assigns an AID and optionally renames with AID prefix.

get_asset

Pull an asset from the library by AID.

Using with Remotion

Generated files land in public/ and work directly with staticFile():

import { Img, Video, Audio, staticFile } from "remotion";

<Img src={staticFile("hero.png")} />
<Video src={staticFile("intro.mp4")} />
<Audio src={staticFile("bgm.mp3")} />

Airtable Integration (Optional)

Connect Airtable to auto-track every generated asset with a unique AID.

Setup

1. Create an Airtable Personal Access Token with data.records:read, data.records:write, and content:manage scopes
2. Create a base with a table (default name: "Assets") with these fields:

3. Set the env vars: AIRTABLE_API_KEY, AIRTABLE_BASE_ID, and optionally AIRTABLE_TABLE_NAME

Without Airtable configured, everything works normally — generation tools save to public/ and asset library tools return a "not configured" message.

Local Development

git clone https://github.com/stephengpope/remotion-media-mcp.git
cd remotion-media-mcp
npm install
npm run build

# Run locally
KIE_API_KEY=your-key node dist/index.js

# Add your local build to Claude Code
claude mcp add remotion-media \
  -e KIE_API_KEY=your-key \
  -- node $PWD/dist/index.js

License

MIT