Steal

Someone's edit works and you want it on your own videos. This skill measures their edit, writes it down as a style, and proves it on one of your takes before you use it.

Output:

• video-agent/styles/<name>/style.json: the style the engine reads, used by /short-editing <take> style <name>.
• video-agent/styles/<name>/STYLE.md: what was measured and the rules the engine can't hold.
• video-agent/styles/<name>/stills/ (key frames of the original) and assets/ (its sound effects and logos, copied into every short that uses it).
• video-agent/edits/short-<slug>-<name>/output/short-<slug>-<name>.mp4: the test video.

Read references/learnings.md before you start. Each rule in it comes from a copy that got rejected.

How to run it

• Preflight: ffmpeg -version && yt-dlp --version. Missing: /studio-setup tools.
• Talk in the creator's language. Say what you do in one line before each long step.
• <name>: the creator's handle in lowercase with dashes, unless the user names the style.
• Scratch files go in video-agent/.tmp/steal-<name>/ (W below).

1. Get the reel

• A local file: use it.
• A YouTube link: yt-dlp -o "W/reel.%(ext)s" "<url>".
• Instagram or TikTok: try yt-dlp first. If it is blocked (login wall), and Treg is set up (treg balance answers), treg catalog search "download an Instagram reel video file", then treg catalog get <id> for the price, then treg call <id> --query url=<url> and download the video URL it returns. Otherwise ask the user to save the video and give you the file.
• When you can, study 2 or 3 reels of the same format. One reel can be the exception.

2. Measure

Do all three passes. A style copied from the frames alone misses half of it: the sound and the rhythm.

Faster with sub-agents. If your agent can run sub-agents (Claude Code, Codex), make the sheets and cuts.txt yourself (first block below), then hand Frames, Sound and Zooms to three sub-agents at the same time. Give each one the absolute paths, its section copied whole, and one file to write: W/notes-frames.md, W/notes-sound.md, W/notes-zooms.md, numbers and timestamps only, never playing audio aloud. Look at the sheets yourself while they work: you judge the look, they bring the numbers. Then write the style from their notes. Without sub-agents, do the passes one after the other.

Frames and cuts

ffmpeg -v error -y -i W/reel.mp4 -vf "fps=2,scale=270:480,tile=8x3:padding=4:color=black" W/sheet-%02d.jpg
ffmpeg -hide_banner -i W/reel.mp4 -vf "scdet=threshold=12" -an -f null - 2>&1 | grep -o "scd.time: [0-9.]*" | cut -d' ' -f2 > W/cuts.txt
ffprobe -v error -show_entries format=duration -of csv=p=0 W/reel.mp4

• Each sheet holds 24 tiles at 2 per second. Sheet N, tile i (from 0) shows second 12(N-1) + i/2.
• Seconds per shot = duration / (number of lines in cuts.txt + 1).

Read EVERY sheet and write down:

• Layouts: full camera, split screen (where the seam sits), flat color with text, cut-out person on a color, B-roll, screenshots.
• Captions, per layout: font feel (condensed, rounded, serif), case, weight, size, where they sit on the body, how a word appears, which word gets emphasised and how.
• Boards and images: type, border, size, when they land (on the word that names them?), whether they stack.
• Hook (first 2 s), CTA (last 3 s), color grade (bright and flat, or contrasty and dark).

Save 6 to 10 key stills to video-agent/styles/<name>/stills/:

ffmpeg -v error -y -ss <seconds> -i W/reel.mp4 -frames:v 1 video-agent/styles/<name>/stills/t<seconds>.jpg

Sound

ffmpeg -v error -y -i W/reel.mp4 -lavfi "showspectrumpic=s=3000x500:legend=0:scale=log:stop=11000" W/spectrum.png

• The image is 3000 px wide, so a cut at t seconds sits at x = 3000 × t / duration. Read it next to cuts.txt.
• A bright thin vertical stroke high up = a click. Wide and low = a boom. A rising smear = a whoosh. A dark gap across the steady music band = the beat drops out.
• Zoom on one cut to see its hit: ffmpeg -v error -y -ss <cut - 1> -t 2 -i W/reel.mp4 -lavfi "showspectrumpic=s=1000x500:legend=0:scale=log:stop=11000" W/cut-<n>.png.
• Beat: count the regular low strokes over 10 s on the spectrogram (or tap along while it plays): strokes per 10 s × 6 = BPM.

Music level under the voice (optional, needs ELEVENLABS_API_KEY in video-agent/.env, uses a few credits, say so first):

set -a; . video-agent/.env; set +a
ffmpeg -v error -y -i W/reel.mp4 -vn -ac 1 -ar 44100 W/full.wav
curl -s -X POST https://api.elevenlabs.io/v1/audio-isolation -H "xi-api-key: $ELEVENLABS_API_KEY" -F audio=@W/full.wav -o W/voice.mp3
ffmpeg -v error -y -i W/voice.mp3 -ac 1 -ar 44100 W/voice.wav
for f in full voice; do ffmpeg -v error -i W/$f.wav -af "asetnsamples=882,astats=metadata=1:reset=1,ametadata=print:key=lavfi.astats.Overall.RMS_level:file=-" -f null - | grep -o "RMS_level=.*" | cut -d= -f2 > W/rms-$f.txt; done
paste W/rms-voice.txt W/rms-full.txt | awk '$1<-40{print $2}' | sort -n | awk '{a[NR]=$1}END{print "music while the voice is silent:", a[int(NR/2)+1], "dB over", NR/50, "s"}'
awk '$1>-30' W/rms-voice.txt | sort -n | awk '{a[NR]=$1}END{print "voice while talking:", a[int(NR/2)+1], "dB"}'

• These commands measure the level every 20 ms. The music is the full mix in the moments the isolated voice is silent.
• Music under the voice = the voice level minus the music level.
• Under 1 s of silence (an edit with every pause cut) is too little to trust: judge it on the spectrogram.
• Never subtract the isolated voice from the mix: it is not in phase with the original, so the result comes out louder than the voice.
• Without a key, judge the level on the spectrogram.

Write down where the music starts and which event gets which sound: cut, board, zoom, impact.

Zooms

On the sheets, measure the face height as a share of the frame height on 10 tiles or so, and list the jumps in face size from one tile to the next:

• a jump at a cut = a change of framing between two shots
• a jump inside a shot = a punch-in
• a slow growth over a shot = a push

Write the range (e.g. face 12 to 25 % of the frame), the punch-ins per 10 s, and whether there is a push. These numbers are the ceiling for your copy.

3. Write the style

video-agent/styles/<name>/STYLE.md, in the user's language, with the same sections as short-editing's style.md. Each line gives a value and the measurement behind it:

# Style: <name>

Copied from <link>, measured on <date>. Applied on top of short-editing's style.md: what is not here stays as there.

## Cuts
- hard cut every 1.4 s on average (35 cuts in 50 s); no fade, no transition
## Framing
- face 12 to 25 % of the frame; punch-in 1.1 at most, about every 4 s, held until the next cut; slow push 0.6 %/s
## Captions
- ...
## Boards and images
- ...
## Sound
- boom bap beat ~90 BPM from the hook impact, 12 dB under the voice, drops out 0.3 s on punch-ins
- a dry high click on most cuts (assets/sfx/cut.mp3), a pop on each board (assets/sfx/pop.mp3)
## Grade
- ffmpeg: eq=contrast=1.08:saturation=0.9,vignette=PI/5,unsharp=5:5:0.4
## Left out
- the AI video clips: replaced by your own footage or a board
## How short-editing applies it
- what goes in each spec by hand (an HTML component, a layout key, a B-roll moment)

Then write video-agent/styles/<name>/style.json with the closest values the engine has, every key from short-editing's style-schema.md: colors, font / display, captions (26 looks: stack for big condensed words on the camera, band, dim, rise, hard...), paper (bg: 'flat' for a solid color background), layout.seam, cuts: 'hard', zoom.drift, pace.speed, grade (an ffmpeg chain), sfx.kit (the cut, pop and punch files of assets/sfx/), music.dropOn, and the rules the engine can't hold as notes. The poster look of assets/looks.json is a measured copy of a hard-cut creator: start from it when the reel cuts hard. Logos the reel shows go in assets/logos/.

Sound effects (in video-agent/styles/<name>/assets/sfx/): take them from the library that ships with short-editing, never generate them by default.

• Pick the closest file in <short-editing>/assets/sfx/ (list and uses in short-editing's references/sfx.md): zhhh, impact-boom, riser, glass-shatter, pop-back cover hook transitions, reveals and payoffs. Copy them into the style's assets/sfx/ and point sfx.kit / sfx.files / the notes at them.
• The engine's own sounds (thud, pop, tap, bell, tom in sfx.extra) cover soft hits with no file at all.
• Only when the original has a sound none of these comes close to, and the creator agrees (it uses their credits), generate 2 or 3 takes with ElevenLabs and keep the closest:

  curl -s -X POST https://api.elevenlabs.io/v1/sound-generation -H "xi-api-key: $ELEVENLABS_API_KEY" -H "content-type: application/json" -d '{"text":"single dry short high click, tight transient, no reverb","duration_seconds":0.5}' -o video-agent/styles/<name>/assets/sfx/cut-a.mp3

Images: a picture the creator puts on screen (a product, a screenshot) becomes an img item on a board; generate one with /codex-images only when they have none, never a generated face.

Logos: every app named on screen shows its real logo, never a letter in a circle.

4. Test it

1. Pick a 10 to 20 s take of the user from video-agent/takes/.
2. Run /short-editing <take> style <name>. init snapshots the style and copies its assets/; the spec adds what STYLE.md asks per short (layout keys, punch-ins, boards on the word that names them).
3. The slug is <take slug>-<name>.

5. Compare before showing

Put the original on top and the copy below, at the same seconds:

ffmpeg -v error -y -i W/reel.mp4 -i video-agent/edits/short-<slug>/output/short-<slug>.mp4 -filter_complex "[0:v]fps=1,scale=216:384,trim=end_frame=15,tile=15x1[a];[1:v]fps=1,scale=216:384,trim=end_frame=15,tile=15x1[b];[a][b]vstack" -frames:v 1 W/compare.jpg

Go tile by tile:

• type size and weight
• board and logo size (they fill the space like in the original)
• grade
• no fade where the original cuts hard
• captions off the mouth
• nothing upside down

Run the spectrogram on the copy too: the hits must land on the cuts, the beat must sit at the same level under the voice. Measure the face size on its sheets: no more zooms than the original, the face never bigger.

With sub-agents, give the compare sheet, the key stills and the list above to a fresh-eyes sub-agent while you check the sound and the zooms: it reports every tile that differs. Fix every gap before showing. The user should never be the one to spot it.

6. Deliver

Open the test video, then tell the user in a few lines:

• what was measured (seconds per shot, BPM, music level, zooms)
• what was copied, and what was left out and why
• where the style lives

From now on, /short-editing <take> style <name> uses it. Each time the user rejects something, write it in STYLE.md (and notes of style.json). When the lesson is general, add it to references/learnings.md.