behind-the-person — Claude Code skill for graphics behind a speaker
A Claude Code skill edits vertical talking-head clips with graphics layered behind the speaker. Run /behind-the-person or ask for visuals behind a person.
MIT ★ 94 updated 2026-10-30
/plugin marketplace add artemnovitckii/claude-editing-skills
/plugin install editing-skills@claude-editing-skillsWhat behind-the-person does
The skill builds a 9:16 talking-head video with the speaker cut out in front of a panel. The panel can show a post grid, a bar chart or a screenshot; large words appear on the speaker’s chest in sync with speech. An optional profile card counts up and changes its button label.
Python scripts in the kit handle transcription, frame selection and rendering from a JSON spec. Transcription requires GROQ_API_KEY; ffmpeg and Node are also needed. On first use, setup.sh --cutout creates the Python environment if it is missing.
Cutout works best with one person, framed from the waist up against a plain background. The panel should use the user’s own posts, analytics or screenshots, and its count must match a number they gave.
What it helps with
- place graphics behind a talking-head speaker
- sync large on-screen words with speech
- show posts or a chart behind the speaker
- add a profile card with a rising count
Example prompts for the agent
- Edit this clip with graphics behind me using
/behind-the-person. - Put a chart behind the speaker and time the on-screen words to the speech.
- Use my posts behind the person and add a profile card with a follower count.
How to install behind-the-person
1Plugin marketplace
/plugin marketplace add artemnovitckii/claude-editing-skills/plugin install editing-skills@claude-editing-skillsThis installs the whole editing-skills plugin with all its skills
2Skills CLI (Vercel)
npx skills add artemnovitckii/claude-editing-skills --skill behind-the-person -a claude-code3Manually: put the files in place
~/.claude/skills/behind-the-person/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/behind-the-person/
1Skills CLI (Vercel)
npx skills add artemnovitckii/claude-editing-skills --skill behind-the-person -a codex2Manually: put the files in place
.agents/skills/behind-the-person/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add artemnovitckii/claude-editing-skills --skill behind-the-person -a cursor2Manually: put the files in place
.cursor/skills/behind-the-person/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install artemnovitckii/claude-editing-skills skills/behind-the-person2Manually: put the files in place
.github/skills/behind-the-person/SKILL.md1Skills CLI (Vercel)
npx skills add artemnovitckii/claude-editing-skills --skill behind-the-person -a gemini-cli2Platform CLI
gemini skills install https://github.com/artemnovitckii/claude-editing-skills --path skills/behind-the-personManually: put the files in place
.devin/skills/behind-the-person/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about behind-the-person
How do I install behind-the-person in Claude Code?
Add the marketplace with /plugin marketplace add artemnovitckii/claude-editing-skills, then install it with /plugin install editing-skills@claude-editing-skills.
What does behind-the-person do?
It edits a vertical clip so the cut-out speaker sits in front of a panel showing posts, a chart or a screenshot. It can also time words to speech and add a counting profile card.
What does behind-the-person need to run?
Transcription needs GROQ_API_KEY; the workflow also needs ffmpeg and Node. First use may run setup.sh --cutout to create the Python environment.
Source text
SKILL.md · MIT
---
name: behind-the-person
description: Edits a talking-head clip so the graphics sit BEHIND the speaker. The speaker is cut out and stays in front of a panel (a grid of posts, a bar chart or a screenshot), big words land on the chest as they are said, and a profile card counts a number up. Use when the user runs /behind-the-person or asks for "visuals behind me", "graphs behind the person" or a professional depth effect on a 9:16 clip.
---
# /behind-the-person
```
┌──────────────────────┐
│ ╭──────────────────╮ │
│ │ grid / chart / │ │ the panel, BEHIND the head
│ │ screenshot head │ │
│ ╰──────── ( ) ─────╯ │
│ 10,000 │ big chest words, one beat at a time
│ (o) Name +10,000 [Follow] │ profile card, counts up, button flips
└──────────────────────┘
```
## First use
This skill's folder is the "Base directory" Claude Code shows when it loads it.
```bash
SKILL_DIR="<base directory of this skill>" # the path Claude Code printed as "Base directory for this skill"
KIT=$(python3 -c 'import os,sys;print(os.path.realpath(os.path.join(sys.argv[1],"..","..","kit")))' "$SKILL_DIR")
PY="$KIT/.venv/bin/python"; [ -x "$PY" ] || bash "$KIT/setup.sh" --cutout
```
`GROQ_API_KEY` must be set to transcribe. Needs ffmpeg and Node.
## Steps
1. **Clip.** The user's 9:16 talking head. The cutout works best on one person, waist up, on a plain background.
2. **Words.** Run `"$PY" "$KIT/reel.py" words <video>`. It prints one sentence per line with `[ends …]` (when its last word ends). Pick a 3 to 6 second stretch of whole sentences with a concrete claim: `start` is the first word's time and `end` is the `[ends …]` time. A claim that holds a number works best ("this post alone got me over 10,000 followers").
3. **Framing.** Run `"$PY" "$KIT/reel.py" frame <video> <t> frame.jpg` and look at it.
- The panel spans y 150 to 1020 by default. The head should cover its lower middle.
- The chest words sit at `chest_top` (1105) and the card at `card_top` (1235). Both should land on the torso, under the chin.
4. **Spec.** Write `<clip>.behind-the-person.json` (below). Paths in a spec are relative to the spec file. Times are seconds from `start`, or spoken words (`"followers"`, `"this#2"`, `"over+0.2"`).
5. **Build.** Run `"$PY" "$SKILL_DIR/build.py" <spec.json>`. The cutout takes about 0.1 s per frame.
6. **Check.** Look at `<out>.sheet.jpg`:
- The chest words don't cover the mouth.
- The panel reads around the head.
- The card is fully visible.
- Nothing sits in the red zones.
7. **Hand over** the mp4 path.
## Spec
```json
{
"video": "my-clip.mp4", "start": 9.4, "end": 13.1,
"panel": {
"type": "grid", "images": "my-posts/*.jpg", "tabs": ["Posts", "Reels", "Tagged"],
"highlight": {"index": 0, "at": "this", "slides": ["post-2.jpg", "post-3.jpg"], "swipe_at": ["carousel", "alone+0.2"],
"label": "10K+ followers", "label_at": "10,000"}
},
"chest": [
{"text": "for example", "at": 0.24}, {"text": "this carousel", "at": "this"}, {"text": "alone", "at": "alone"},
{"text": "got me over", "at": "got"}, {"text": "{count}", "at": "10,000"}, {"text": "followers", "at": "followers"}
],
"card": {"avatar": "me.jpg", "name": "Your Name", "label": "new followers", "count": "10,000", "at": "got",
"count_at": "10,000", "button": "Follow", "pressed": "Following", "press_at": "followers"}
}
```
The panel takes one of three `type`s:
- **`grid`**: an Instagram-style profile grid with 12 cells, filled with your images on repeat, that scrolls slowly. `highlight` outlines one cell, swipes through its `slides` and pops a `label` chip.
- **`bars`**: `{"type": "bars", "title": "Views per post", "bars": [{"label": "Reels", "value": 12000, "show": "12K"}, ...], "highlight": 2, "at": 0.3}`. The bars grow in turn. Put the bar that matters at the side, because the middle sits behind the head.
- **`image`**: `{"type": "image", "src": "screenshot.png", "scroll": 200}`, a screenshot that pans upward.
The other fields:
- **`chest`**: the big words, each shown from its `at` until the next one. `{count}` counts up to `card.count` (`"10,000"`, `"4:12"` or `"250"`).
- **`card`**: optional. It counts up with the chest and the button flips at `press_at`.
## Taste
- Chest words are 1 to 3 words: the beat, not the sentence. Land each one on its word.
- The panel's content must be real: the user's own posts, analytics or screenshots. Never invent their numbers. The count must be one they said.
- Keep it calm. One panel, one highlight and one card is enough. The motion comes from the speaker standing in front of it.
© Artem Novitckii · MIT · artemnovitckii/claude-editing-skills