MCP Integration

The Model Context Protocol (MCP) enables AI assistants to create and edit videos using natural language. Instead of writing JSON specifications, describe what you want and the AI generates it.

Quick Start

Add the Shotstack MCP server to your AI assistant. For Claude Desktop, add to your config file:

macOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json

Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "shotstack": {
      "command": "npx",
      "args": ["-y", "@shotstack/shotstack-mcp-server"],
      "env": {
        "SHOTSTACK_API_KEY": "your_api_key"
      }
    }
  }
}

Restart Claude Desktop afterwards.

danger

The MCP server currently only supports production API keys. Sandbox/stage keys will not work.

Get your production API key from dashboard.shotstack.io, then try:

Create a 10-second video with 'Hello World' text centered on screen

See Platform Setup for other AI assistants.

Available Tools

Tool	Description
render_video	Create a video from a JSON specification
get_render_status	Check render progress and get output URL
create_template	Save a reusable video template
list_templates	List all saved templates
get_template	Retrieve a template's structure
render_template	Render a video from a template with merge fields
delete_template	Remove a template
inspect_media	Get metadata about a media file

See Tools Reference for parameters and examples.

Learn More

• Platform Setup — Configuration for each AI assistant
• Tools Reference — Complete tool documentation
• Best Practices — Tips for effective AI video generation