Turn a folder of raw footage into a finished edit with an agent skill
by PaperChaseWebb
video-use is an open-source skill from the Browser Use team. You drop raw takes in a folder and ask Claude Code or Codex for an edit. It cuts filler words, grades color, adds subtitles and animated overlays, then checks its own render before showing you.
- Price
- Free
- Difficulty
- Beginner
- Setup
- 15 min
- Format
- .md file
turn-a-folder-of-raw-footage-into-a-finished-edit-with-an-agent-skill.md · 2 KB · direct download, no signup
What you end up with
The approval step before rendering and the self-check after are what make it usable.
What you get
- A plain-language summary of the workflow
- A fill-in prompt template
- What to watch out for
- A link to the original source
Who it is for
Builders who already use AI agents and want a proven workflow they can copy into their own project.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: Plain Markdown, so it also pastes into other agents such as Grok Bot, Hermes, OpenClaw or Muse. We list those as compatible by format only, not as tested.
Tools needed
Cost to run: Free with the agent plan you already use.
Time to result: One working session
How it works
Install ffmpeg and set up the skill by following its install file, wit
Install ffmpeg and set up the skill by following its install file, with the agent asking you for the API key
Expected: Done before you move to the next step.
Put raw takes for one video in a fresh folder
Put raw takes for one video in a fresh folder
Expected: Done before you move to the next step.
Ask for the edit and read the plan before you approve it
Ask for the edit and read the plan before you approve it
Expected: Done before you move to the next step.
Review the cut at every join, then ask for specific changes
Review the cut at every join, then ask for specific changes
Expected: Done before you move to the next step.
Keep the project notes file so the next edit starts from your choices
Keep the project notes file so the next edit starts from your choices
Expected: Done before you move to the next step.
Sample prompt from the pack
Apply this workflow to [your project]: 1. Install ffmpeg and set up the skill by following its install file, with the agent asking you for the API key 2. Put raw takes for one video in a fresh folder 3. Ask for the edit and read the plan before you approve it 4. Review the cut at every join, then ask for specific changes 5. Keep the project notes file so the next edit starts from your choices Stop and ask me before anything irreversible: deploys, payments, sends or deletes.
Adapted by PaperChaseWebb. Not the original author's prompt.
Free preview of the file
# Turn a folder of raw footage into a finished edit with an agent skill PaperChase agent use case (free feed entry), version 1.0. Source: [Browser Use team (browser-use/video-use on GitHub)](https://github.com/browser-use/video-use), 2026-10-26. ## What they did The model never watches the video. It reads a transcript with word-level timings and other extracted data, proposes an edit plan and waits for your OK, then builds the cut with ffmpeg: filler words and dead air removed, short audio fades at every cut, color grade, burned-in subtitles and overlays made with HyperFrames, Remotion or Manim by parallel subagents. It reviews the result at each cut point and keeps notes in a project file for the next session. MIT licensed. ## Worth knowing The approval step before rendering and the self-check after are what make it usable. It needs ffmpeg and an ElevenLabs key, so footage goes to a third party for transcription; keep private recordings out of it. Star counts and features come from the repo and promoters.
Worth knowing
- The approval step before rendering and the self-check after are what make it usable. It needs ffmpeg and an ElevenLabs key, so footage goes to a third party for transcription; keep private recordings out of it. Star counts and features come from the repo and promoters.
- Figures are as reported by the original author and were not re-measured by us.
Questions
Where does this come from?
A public post credited to Browser Use team (browser-use/video-use on GitHub). Source: https://github.com/browser-use/video-use
Is it free?
Yes. Feed entries are free to read and download.
Details
Version 1.0 · Updated Oct 26, 2026 · Markdown version of this page
License: Personal and commercial use. Credit the original source when you share the workflow. Download terms
Related

Video Production with AI Agents: lanes, prompts, skills and render checks
A Markdown playbook for making finished videos with Claude Code or Codex: pick a lane (draw in code, edit footage, script 3D or generate), plan with a shot list, render test frames, and make the agent check every render before you see it.

Check every render before you trust it (free sample)
One rule, one skill and four ffmpeg commands that make Claude Code or Codex look at its own video output (frames, cuts, loudness, captions) before it tells you it is done.