Free during launch: every app and tool is unlimited — no account, no card. Purchases open soon (then $9 once). Download · Free browser tools
GUI Hub
For Claude, Cursor, VS Code & other MCP clients

Can Claude watch a video or screen recording?

Not on its own. AI assistants — including coding agents like Claude Code and Codex — can read text and look at images, but they can’t watch a video or hear its audio. With the GUI Hub Essentials MCP server installed, Claude (Desktop or Code), Cursor, VS Code and other MCP clients can watch a video (screenshots + transcript) and answer questions about it locally: the assistant calls the watch_video tool, which runs FFmpeg + Whisper on your machine. Files are not uploaded.

Set it up (2 minutes)

  1. Claude Code / Cursor / VS Code / Windsurf: add the MCP server
    npx -y https://guihub.cc/mcp.tgz --product=essentials
    Claude Code: claude mcp add gui-hub-essentials -- npx -y https://guihub.cc/mcp.tgz --product=essentials
  2. Claude Desktop: download the one-file extension and double-click it.
  3. No AI? GUI Hub Essentials is also a point-and-click desktop app.

Agent with its own computer?

Grok Bot, ChatGPT dots, Claude Duet, OpenClaw, Manus or Codex can run it directly — no MCP needed:

npx -y https://guihub.cc/cli.tgz run watch_video --input /path/to/input.mp4 --outDir /path/to/folder

Options: npx -y https://guihub.cc/cli.tgz info watch_video · skill file: SKILL.md · more for agents

Then ask

“Watch demo.mp4 and write step-by-step instructions for exporting a report, with the right screenshot for each step.”

What the AI gets

watch_video — Let the AI 'watch' a video on the user's computer, offline: saves a screenshot every `every` seconds into `outDir` (frame_0001.jpg = 0:00, frame_0002.jpg = `every` seconds later, and so on; at most 300) plus transcript.txt with timestamped speech ([hh:mm:ss] lines, Whisper). Then open the screenshots and the transcript yourself to answer questions about the video, summarise it, write step-by-step docs or a tutorial from a screen recording, find when something is shown or said, or pick timestamps for clips. Use `every` 2 for fast screen recordings, 5 for normal videos, 15-30 for long lectures or meetings. Silent videos are fine (the transcript then says no speech was found).

InputType
input*string
everystring (2, 5, 15, 30)
languagestring (, en, es, fr, de, it)
outDir*string

Cost

Free to install; each tool has 3 free uses. Then $9 once for all of GUI Hub Essentials (no subscription). The tool tells the assistant when free uses run out and how to unlock.

Powered by FFmpeg + Whisper (GPL (FFmpeg static build), Apache-2.0/MIT (Whisper via sherpa-onnx)), the official release, verified by SHA-256. GUI Hub is independent and not affiliated with FFmpeg + Whisper, Anthropic, Cursor or Microsoft.

Other things your AI can do with GUI Hub