# video-edit

video-edit turns a raw clip into a captioned video: transcribes the speech, lets you fix the text, then renders 16:9 or 9:16 from the same file.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf (Devin Desktop), Claude.ai · ★ 301 repo stars
- Author: hoodini
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/hoodini/ai-agents-skills/tree/HEAD/skills/video-edit/SKILL.md
- Updated: 2026-10-23
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/hoodini-video-edit

## What it does

The skill runs the whole job. `ffprobe` reads the source, `npx hyperframes init` scaffolds a project, `ffmpeg` pulls the audio, and `transcribe.py` runs Whisper with `large-v3` as the model. English, Hebrew and any language Whisper handles are covered.

Then it stops on purpose. A local review server opens a page where you correct misheard words and approve; the server writes `transcript_review.txt` and exits. Only after that does `apply_review.py` push word timings back into the transcript so captions stay in sync, and the render starts: liquid-glass caption pills, blob backgrounds, morph wipes.

You need `ffmpeg` and `ffprobe`, Node for the HyperFrames CLI, Python for the bundled scripts. On Windows CUDA usually fails without cuDNN, so the script drops to CPU int8. The vertical cut reuses the same 16:9 source: centered footage strip, blurred backdrop, blobs, a caption pill sized for portrait. No reshoot.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add hoodini/ai-agents-skills --skill video-edit -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/video-edit/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/video-edit/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add hoodini/ai-agents-skills --skill video-edit -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/video-edit/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add hoodini/ai-agents-skills --skill video-edit -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/video-edit/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install hoodini/ai-agents-skills skills/video-edit
```

2. **Manually: put the files in place**
- File: `.github/skills/video-edit/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add hoodini/ai-agents-skills --skill video-edit -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/hoodini/ai-agents-skills --path skills/video-edit
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/video-edit/SKILL.md` — The legacy .windsurf/skills/ path still works

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/hoodini/ai-agents-skills/tree/HEAD/skills/video-edit/SKILL.md

## When to use it
- caption a screen recording or a talk without typing subtitles by hand
- fix Hebrew mishears in a transcript before the long render
- cut a 9:16 reel from a horizontal master
- put text behind the speaker with background removal
- repeat a captioned look on a new clip

## Example prompts
- Caption `intro.mp4` in Hebrew and render a 9:16 version for Reels.
- Fix the misheard words in the transcript, then re-render.
- Do the same captioned style as the last video, source is `demo.mov`.

## Similar
- [viral-video-animated-captions](https://affpapa.org/en/nevibecode/skills/viral-video-animated-captions.md) — skill · 196 installs
- [talking-head-recut](https://affpapa.org/en/nevibecode/skills/talking-head-recut.md) — skill · ★ 328 repo stars
- [short-form-video](https://affpapa.org/en/nevibecode/skills/notivn-short-form-video.md) — skill · ★ 147 repo stars
- [reel-editor](https://affpapa.org/en/nevibecode/skills/reel-editor.md) — skill · 875 installs
- [ffmpeg-captions-subtitles](https://affpapa.org/en/nevibecode/skills/ffmpeg-captions-subtitles.md) — skill · 634 installs
- [stage-edit](https://affpapa.org/en/nevibecode/skills/stage-edit.md) — skill · ★ 350 repo stars

## Other versions with the same name
- [video-edit · prime-skills](https://affpapa.org/en/nevibecode/skills/video-edit.md) — 450700 installs
- [video-edit · calesthio](https://affpapa.org/en/nevibecode/skills/calesthio-video-edit.md) — 1600 installs
- [video-edit · video-db](https://affpapa.org/en/nevibecode/skills/video-db-video-edit.md) — 177 installs
- [video-edit · 0xsero](https://affpapa.org/en/nevibecode/skills/0xsero-video-edit.md) — ★ 92 repo stars
- [video-edit · yildizdikme](https://affpapa.org/en/nevibecode/skills/yildizdikme-video-edit.md) — ★ 57 repo stars

From the same set: [hoodini/ai-agents-skills](https://affpapa.org/en/nevibecode/by/hoodini/ai-agents-skills.md) — 34 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
