mcp-video MCP Server

<strong>The video editing MCP server for AI agents and humans.</strong> 26 tools. 3 interfaces. Rich CLI. Purpose-built for AI agents.

Quick Start • Tools • Python API • CLI • Timeline DSL • Templates • Roadmap • Contributing

What is mcp-video?

mcp-video is an open-source video editing server built on the Model Context Protocol (MCP). It gives AI agents, developers, and video creators the ability to programmatically edit video files.

Think of it as ffmpeg with an API that AI agents can actually use. Instead of memorizing cryptic command-line flags, an agent calls structured tools with clear parameters and gets structured results back — or a human uses the rich CLI with spinners, tables, and error panels.

The Problem It Solves

AI agents can write code, analyze documents, and browse the web — but they can't edit video. Existing video editing tools are either:

• GUI-only (Premiere, DaVinci, CapCut) — agents can't use them
• Raw FFmpeg wrappers — require memorizing hundreds of flags
• Cloud APIs (Render, Bannerbear) — expensive, slow, vendor lock-in

mcp-video bridges this gap. It's a local, fast, free video editing layer that any AI agent can use through a standard protocol.

Three Ways to Use It

Install

Prerequisites

FFmpeg must be installed on your system:

# macOS
brew install ffmpeg

# For full text overlay support (drawtext filter):
# brew install freetype harfbuzz
# brew reinstall --build-from-source ffmpeg
# Verify: ffmpeg -filters | grep drawtext

# Ubuntu/Debian
sudo apt install ffmpeg

# Windows
# Download from https://ffmpeg.org/download.html

Installation

pip install mcp-video

Or with UVX (no install needed):

uvx mcp_video

Quick Start

1. As an MCP Server (for AI agents)

Install FFmpeg first, then pick your client:

Claude Code:

claude mcp add mcp-video -- pip install mcp-video && mcp-video --mcp

Claude Desktop — add to your claude_desktop_config.json:

{
  "mcpServers": {
    "mcp-video": {
      "command": "uvx",
      "args": ["mcp_video"]
    }
  }
}

Cursor — add to your .cursor/mcp.json:

{
  "mcpServers": {
    "mcp-video": {
      "command": "uvx",
      "args": ["mcp_video"]
    }
  }
}

Any MCP client — if installed via pip, the command is just mcp-video.

Then just ask your agent: "Trim this video from 0:30 to 1:00, add a title card, and resize for TikTok."

2. As a Python Library

from mcp_video import Client

editor = Client()

# Get video info
info = editor.info("interview.mp4")
print(f"Duration: {info.duration}s, Resolution: {info.resolution}")

# Trim a clip
clip = editor.trim("interview.mp4", start="00:02:15", duration="00:00:30")

# Merge clips with transitions
video = editor.merge(
    clips=["intro.mp4", clip.output_path, "outro.mp4"],
    transitions=["fade", "dissolve", "fade"],
)

# Add text overlay
video = editor.add_text(
    video=video.output_path,
    text="EPISODE 42: The Future of AI",
    position="top-center",
    size=48,
)

# Add background music
video = editor.add_audio(
    video=video.output_path,
    audio="music.mp3",
    volume=0.7,
    fade_in=2.0,
    fade_out=3.0,
)

# Resize for TikTo