video-use
GitHub

Open source · MIT

Edit videos with coding agents.

Drop raw footage in a folder, chat with Claude Code or Codex, get final.mp4 back. Talking heads, tutorials, montages, interviews. No presets, no timeline.

Cloud-hosted video-use is coming. Upload footage, get the cut back. No local setup.

Claude Code editing a folder of raw takes with video-use: inventory, transcript scan, strategy, then render and self-check

What it does

The agent reads the video, it never watches it.

Every take becomes a word-level transcript with speakers and audio events, packed into one ~12KB file. The agent pulls a filmstrip and waveform only at decision points. Same idea as giving a browser agent the DOM instead of a screenshot.

Frame dumping: 30,000 frames × 1,500 tokens = 45M tokens.
video-use: 12KB of text + a handful of PNGs.

## C0103  (duration: 43.0s, 8 phrases)
  [002.52-005.36] S0 Ninety percent of what a web agent does is completely wasted.
  [006.08-006.74] S0 We fixed this.

Run it yourself today

Needs ffmpeg and an ElevenLabs key for transcription. Your agent handles the rest.

Paste into Claude Code, Codex, or any agent with a shell
Set up https://github.com/browser-use/video-use for me.

Read install.md first to install this repo, wire up ffmpeg, register the skill with whichever agent you're running under, and set up the ElevenLabs API key - ask me to paste it when you need it. Then read SKILL.md for daily usage, and always read helpers/ because that's where the editing scripts live. After install, don't transcribe anything on your own - just tell me it's ready and wait for me to drop footage into a folder.