VidContext MCP Server

Give any MCP-compatible AI agent the ability to analyze video. Install with pip, add your API key, and your AI can extract transcripts, visual scenes, brands, and audio from any video file — as a native tool call.

Quick start

1. 1

   Install the package

   Run pip install vidcontext-mcp in your terminal. Requires Python 3.9 or later.

2. 2

   Get your API key

   Sign up at vidcontext.com/app and go to the Developer page. Copy your API key.

3. 3

   Configure your AI client

   Add the VidContext MCP server to your client's configuration file. See the examples below for Claude Desktop and Cursor.

4. 4

   Start analyzing video

   Ask your AI agent to analyze a video file. It will call the VidContext tool automatically, process the video, and return structured results.

Installation

pip install vidcontext-mcp

# Run manually (useful for testing):
vidcontext-mcp --api-key vc_your_key_here

Claude Desktop configuration

Add this to your Claude Desktop config file (claude_desktop_config.json):

{
  "mcpServers": {
    "vidcontext": {
      "command": "vidcontext-mcp",
      "args": ["--api-key", "vc_your_key_here"]
    }
  }
}

Cursor configuration

The same JSON works for Cursor — save it as .cursor/mcp.json in your project root. See the full Cursor guide for detailed setup.

What your AI agent can do

Frequently asked questions