video-edit — footage to a film, the agent as editor
A folder of clips becomes a trailer, a longer episode, or a film of a whole trip. The tools do the
boring part (shots, quality, beats, speech); the agent picks the story; the person judges. Every
rule below was paid for by a cut someone called boring or broken. If the project has its own
editing skill, that one holds the commands; this one holds the rules.
Workflow
- Analyse once, store beside the sources. Shots, frame quality, beats, speech with word
timings. Write the results as sidecar files next to each clip (
clip.mp4 → clip.analysis.json),
so a month later a long film needs no re-analysis. Take shooting time from container metadata
(ffprobe -show_entries format_tags=creation_time), never from file names: camera clocks drift.
Tools: any shot detector + whisper-family STT; npm i life2film-engine (WASM) covers beats,
shots and frame scores on-device.
- The day as one text file. All speech in shooting order, numbered by clip and sentence. Read
it like a scriptwriter and propose the story before touching a timeline.
- Write the cut as a script, not a timeline. A small JSON: which sentences of which clip are
said, which shot covers which line, which silent walks keep their own sound, what the cold open
is. Render from it. A complaint ("X is mentioned, never shown") becomes a one-line edit.
- Voice as one continuous track, pictures on top. Cut the audio from the same pieces as the
picture, quantised to frames; check that film length equals voice length (drift > 1 frame = bug).
- Look before you render, every time (see Eyes below), then render, then check with
ffprobe
that duration, codec and audio length match the plan.
- Masters per platform from one archival master (
references/delivery.md). Publish private, watch on a
phone, then public.
Editorial rules
- The cutaway shows what is being said, from the same place and time (within minutes of the
line), never from a scene the story has not reached or has left.
- Never cover a shot where lips move. It reads as broken sync. Cutaways go into silent windows.
- A mentioned object must be seen, ideally just before the line about it.
- Cold open, then hook. 8–10 s of the best frames, then one line with a twist. The hook is
hand-picked from a contact sheet: automatic ranking picks sharp and colourful (an empty path
through trees), which is what people scroll past. Fire, water, neon, night, motion toward the
camera, a face with an expression, the film's object.
- Scenes, not best phrases. A conversation goes in whole: lead-in, the talk, the reaction. A
"best line" without its lead-in confuses anyone who did not see what was cut.
- Find promise → payoff pairs in the transcripts ("tomorrow we'll buy…" → the purchase) and
keep both halves; they are the thread of a multi-day film.
- Trim filler words at the ends of a line; mid-line only when the join lands in silence.
- Take something from every clip. The person filmed it, so it mattered. After the cut, list the
clips that did not make it, each with a reason.
- Live music in frame is one continuous span with picture cuts on top. Cut per shot and the beat
restarts at every cut.
- Length follows the material. Cut stammers and repeats, not thoughts. Don't stretch to a target.
Eyes — numbers miss what a glance catches
- Read the whole film as text in order before rendering: does each line make sense to someone
who did not see the cut material?
- A sheet of every cutaway with its line under it. Measured once: 12 of 54 automatic cutaways
were off topic (a frame upside down, a car over a line about a garage).
- A storyboard grid, one frame per shot. Catches "same face ten times" that every metric passed.
- A critic subagent, two passes minimum. Pass one finds repeats and sagging scenes, pass two the
lines cut mid-word at joins.
Sound
Score written to the cut, a hole in it (not ducking) wherever something plays live, live music as
one continuous span, other people's music stripped with Demucs before upload, chapters levelled to
−16 LUFS. Rules, numbers and commands: references/sound.md.
Delivery
One archival master, a file per door: HEVC for upload (36 min = 1.6 GB vs 3.2 GB x264), YouTube in
1440p even from 1080p, h264 for messengers, native vertical untouched or a face-following 9:16
window with burned subtitles for feeds. ffmpeg recipes, the outro card without re-encoding, cover and
archive rules: references/delivery.md.
Gotchas
- Analysis written into a symlinked subset folder ends up beside the links, not the originals —
resolve symlinks before writing sidecars.
- Never leave a rendered film inside the source folder: the next run treats it as one more clip.
- STT turns unknown names into celebrities. Keep a per-trip glossary and fix transcripts once.
- A long trip film is cut from the finished day films (colour, sync, subtitles already fixed), one
file per chapter cached by its inputs, so a critic note re-renders one chapter, not an hour.
- Put the real shooting date in every description; a week in, nobody can tell the days apart.
- Archive sources to cloud storage after the films are out, never mid-edit, with the sidecars.
- Verify on at least one real captured file, not only synthetic ones. A pipeline green on
lavfi test sources can still break on camera files: a 48 kHz silent card joined to a 96 kHz
recording passed every synthetic test and drifted the audio clock 1.39 s at the seam. Keep one
short real clip per camera/phone in the verification set, and check the delivered file itself.
Don't
- Don't trust an automatic hook, cover frame or cutaway list without looking at it.
- Don't re-encode video to change audio, trim, mute or add an outro: copy the stream (
references/delivery.md, "Copy first").
- Don't crop horizontal footage blindly to 9:16.
- Don't chop live music per shot, or duck the score under it.
- Don't render before reading the cut as text.
1---2name: solo-video-edit3description: Edit real footage into a film with the agent as the editor — read the day's speech as text, write the cut as a script, lay cutaways that match the words, score it, strip other people's music, and render masters for YouTube, Reels and messengers with ffmpeg. Use when user says "смонтируй", "сделай ролик из этих видео", "трейлер дня", "влог", "cut this footage", "edit my trip videos", "make a film from these clips", or drops a folder of clips expecting a film. Do NOT use for planning a product promo from scratch (use /video-promo) or for uploading and channel analytics (use /youtube-channel).4license: MIT5metadata:6 source: https://skillmd.com/skills/fortunto2/solo-video-edit7 author: fortunto28 version: "1.0.5"9 openclaw:10 emoji: "🎬"11---1213# video-edit — footage to a film, the agent as editor1415A folder of clips becomes a trailer, a longer episode, or a film of a whole trip. The tools do the16boring part (shots, quality, beats, speech); the agent picks the story; the person judges. Every17rule below was paid for by a cut someone called boring or broken. If the project has its own18editing skill, that one holds the commands; this one holds the rules.1920## Workflow21221. **Analyse once, store beside the sources.** Shots, frame quality, beats, speech with word23 timings. Write the results as sidecar files next to each clip (`clip.mp4` → `clip.analysis.json`),24 so a month later a long film needs no re-analysis. Take shooting time from container metadata25 (`ffprobe -show_entries format_tags=creation_time`), never from file names: camera clocks drift.26 Tools: any shot detector + whisper-family STT; `npm i life2film-engine` (WASM) covers beats,27 shots and frame scores on-device.282. **The day as one text file.** All speech in shooting order, numbered by clip and sentence. Read29 it like a scriptwriter and propose the story before touching a timeline.303. **Write the cut as a script, not a timeline.** A small JSON: which sentences of which clip are31 said, which shot covers which line, which silent walks keep their own sound, what the cold open32 is. Render from it. A complaint ("X is mentioned, never shown") becomes a one-line edit.334. **Voice as one continuous track, pictures on top.** Cut the audio from the same pieces as the34 picture, quantised to frames; check that film length equals voice length (drift > 1 frame = bug).355. **Look before you render, every time** (see Eyes below), then render, then check with `ffprobe`36 that duration, codec and audio length match the plan.376. **Masters per platform** from one archival master (`references/delivery.md`). Publish private, watch on a38 phone, then public.3940## Editorial rules4142- **The cutaway shows what is being said, from the same place and time** (within minutes of the43 line), never from a scene the story has not reached or has left.44- **Never cover a shot where lips move.** It reads as broken sync. Cutaways go into silent windows.45- **A mentioned object must be seen**, ideally just before the line about it.46- **Cold open, then hook.** 8–10 s of the best frames, then one line with a twist. The hook is47 **hand-picked from a contact sheet**: automatic ranking picks sharp and colourful (an empty path48 through trees), which is what people scroll past. Fire, water, neon, night, motion toward the49 camera, a face with an expression, the film's object.50- **Scenes, not best phrases.** A conversation goes in whole: lead-in, the talk, the reaction. A51 "best line" without its lead-in confuses anyone who did not see what was cut.52- **Find promise → payoff pairs** in the transcripts ("tomorrow we'll buy…" → the purchase) and53 keep both halves; they are the thread of a multi-day film.54- **Trim filler words at the ends of a line**; mid-line only when the join lands in silence.55- **Take something from every clip.** The person filmed it, so it mattered. After the cut, list the56 clips that did not make it, each with a reason.57- **Live music in frame is one continuous span** with picture cuts on top. Cut per shot and the beat58 restarts at every cut.59- **Length follows the material.** Cut stammers and repeats, not thoughts. Don't stretch to a target.6061## Eyes — numbers miss what a glance catches6263- **Read the whole film as text** in order before rendering: does each line make sense to someone64 who did not see the cut material?65- **A sheet of every cutaway with its line under it.** Measured once: 12 of 54 automatic cutaways66 were off topic (a frame upside down, a car over a line about a garage).67- **A storyboard grid, one frame per shot.** Catches "same face ten times" that every metric passed.68- **A critic subagent, two passes minimum.** Pass one finds repeats and sagging scenes, pass two the69 lines cut mid-word at joins.7071## Sound7273Score written to the cut, a hole in it (not ducking) wherever something plays live, live music as74one continuous span, other people's music stripped with Demucs before upload, chapters levelled to75−16 LUFS. Rules, numbers and commands: `references/sound.md`.7677## Delivery7879One archival master, a file per door: HEVC for upload (36 min = 1.6 GB vs 3.2 GB x264), YouTube in801440p even from 1080p, h264 for messengers, native vertical untouched or a face-following 9:1681window with burned subtitles for feeds. ffmpeg recipes, the outro card without re-encoding, cover and82archive rules: `references/delivery.md`.8384## Gotchas8586- Analysis written into a symlinked subset folder ends up beside the links, not the originals —87 resolve symlinks before writing sidecars.88- Never leave a rendered film inside the source folder: the next run treats it as one more clip.89- STT turns unknown names into celebrities. Keep a per-trip glossary and fix transcripts once.90- A long trip film is cut from the finished day films (colour, sync, subtitles already fixed), one91 file per chapter cached by its inputs, so a critic note re-renders one chapter, not an hour.92- Put the real shooting date in every description; a week in, nobody can tell the days apart.93- Archive sources to cloud storage **after** the films are out, never mid-edit, with the sidecars.94- **Verify on at least one real captured file, not only synthetic ones.** A pipeline green on95 `lavfi` test sources can still break on camera files: a 48 kHz silent card joined to a 96 kHz96 recording passed every synthetic test and drifted the audio clock 1.39 s at the seam. Keep one97 short real clip per camera/phone in the verification set, and check the delivered file itself.9899## Don't100101- Don't trust an automatic hook, cover frame or cutaway list without looking at it.102- Don't re-encode video to change audio, trim, mute or add an outro: copy the stream (`references/delivery.md`, "Copy first").103- Don't crop horizontal footage blindly to 9:16.104- Don't chop live music per shot, or duck the score under it.105- Don't render before reading the cut as text.