mcp-video

Overview

mcp-video provides a comprehensive video editing toolkit for AI agents, wrapping FFmpeg and other media tools behind structured MCP schemas. It offers tools for trimming, merging, resizing, transcoding videos, adding subtitles, watermarks, effects, and generating AI transcriptions. The server also supports Hyperframes for cinematic video creation and automated repurposing of content across different platforms. With both Python client and CLI interfaces, it enables integration with Claude Code, Cursor, and other AI development environments.

Try asking AI

After installing, here are 3 things you can ask your AI assistant:

• trimtrim(input: str, start: str, duration: str)

  Trim a video to a specific start time and duration

  Try:Trim the video from 1 minute 15 seconds to 45 seconds longTry:Cut a 30-second clip starting at 2 minutes
• subtitlessubtitles(input: str, subtitle_file: str)

  Add subtitles to a video from an SRT file

  Try:Add burned-in subtitles to the video using captions.srtTry:Overlay subtitles on the video
• resizeresize(input: str, aspect_ratio: str)

  Resize a video to a specific aspect ratio

  Try:Make the video vertical for TikTok (9:16 aspect ratio)Try:Convert to square format for Instagram
• release_checkpointrelease_checkpoint(input: str)

  Create quality checkpoints, thumbnails, and storyboards for video review

  Try:Generate a release checkpoint with thumbnails for the videoTry:Create review artifacts for quality control
• ai_transcribeai_transcribe(input: str, output_srt: str)

  Transcribe video audio to an SRT subtitle file

  Try:Generate subtitles from the audio in the videoTry:Create an SRT file with transcription
• normalize_audionormalize_audio(input: str)

  Normalize audio levels in a video

  Try:Make the audio volume consistent throughout the videoTry:Normalize audio levels
• search_toolssearch_tools(query: str)

  Search for tools matching a query string

  Try:Find tools for adding text to videosTry:Search for subtitle-related tools
• scene_detectionscene_detection(input: str)

  Detect scene changes in a video

  Try:Find all the scene breaks in this videoTry:Detect where scenes change
• gridgrid(input: str, layout: str)

  Arrange video clips in a grid layout

  Try:Create a 2x2 grid of four clipsTry:Arrange videos in a 3x3 grid layout
• video_quality_checkvideo_quality_check(input: str)

  Check video quality for issues

  Try:Analyze the video for quality problemsTry:Check for audio or visual issues
• mergemerge(inputs: list[str])

  Merge multiple video files into one

  Try:Combine these three video clips into oneTry:Merge several video files
• animated_textanimated_text(input: str, text: str, position: str)

  Add animated text to a video

  Try:Add 'Welcome!' text that fades in at the topTry:Create animated title at the center

Comparable tools

Installation

# macOS
brew install ffmpeg

# Ubuntu/Debian
sudo apt install ffmpeg

uvx --from mcp-video mcp-video doctor

pip install mcp-video
mcp-video doctor

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

Compare mcp-video with