Getting Started

Video Context MCP is an MCP server that gives AI coding assistants — GitHub Copilot, Cursor, and Claude Code — the ability to understand and analyze video content.

Watch the demo →
Tutorials playlist →

Prerequisites

Node.js 20+ — Required to run the server (download)
A supported MCP client — VS Code + GitHub Copilot Chat, Cursor, or Claude Code
At least one video and one audio API key — See Video Providers and Audio Providers for options; several have free tiers with no credit card required

Install

Install the package globally via npm:

npm install -g video-context-mcp-server

Check Version & Update

# Check the installed version
npm ls -g video-context-mcp-server

# Check if an update is available
npm outdated -g video-context-mcp-server

# Update to the latest version
npm install -g video-context-mcp-server@latest

Supported Input Formats

Format	Example
Local file path	`./demo.mp4`, `/home/user/recordings/clip.mp4`
Windows path	`C:\Users\alice\Videos\demo.mp4`
`file://` URI	`file:///home/user/videos/clip.mp4`
Remote HTTP/S URL	`https://example.com/video.mp4`
YouTube URL	`https://www.youtube.com/watch?v=abc123`
Other platforms	Vimeo, TikTok, Bilibili, etc. — requires Pro

Next Steps

Once installed, configure your MCP client to connect to the server:

VS Code + GitHub Copilot — Create a .vscode/mcp.json in your workspace
Cursor — Project-level or global MCP config
Claude Code — CLI claude mcp add or .mcp.json file
Run via npx — No global install required; self-updates automatically