video-check — скилл для Claude Code: проверка видео
Скилл video-check для Claude Code проверяет готовый ролик перед отправкой или публикацией. Находит дефекты изображения и звука, затем помогает исправить самые заметные.
Apache-2.0 ★ 52 обновлён 2026-11-16
/plugin marketplace add EditAtlas/skills
/plugin install video-check@editatlas-skillsЧто делает video-check
Скилл измеряет размер файла, частоту кадров, формат пикселей и длительность, ищет проблемы изображения и звука. Он создаёт листы кадров с отмеченной безопасной зоной для текста, а отдельный ревьюер без знания замысла перечисляет дефекты с таймкодами и вариантами исправления. Проверка запускается для локального файла после рендера или по просьбе проверить ролик перед отправкой.
Сначала исправляются критичные ошибки, затем ролик рендерится и проверяется снова; цикл ограничен тремя раундами. Для работы нужны Python 3.9+, numpy, Pillow, FFmpeg и ffprobe. Скилл работает офлайн. Скрипт измеряет параметры звука, но оценить разборчивость речи и уместность музыки может только человек.
Какие задачи решает
- найти чёрные и белые кадры, зависания и рывки
- обнаружить щелчки, обрезанный звук и тишину
- сверить длительность, размер и частоту кадров с заданным планом
- получить список дефектов с таймкодами и исправлениями
- повторно проверить ролик после правок
Примеры запросов агенту
- Проверь готовый ролик перед отправкой и перечисли ошибки с таймкодами.
- Посмотри, есть ли дефекты в этом рендере, и предложи исправления.
- Проверь видео после монтажа, исправь самые серьёзные проблемы и запусти проверку повторно.
Как установить video-check
1Через маркетплейс плагинов
/plugin marketplace add EditAtlas/skills/plugin install video-check@editatlas-skillsКоманда ставит плагин video-check целиком, со всеми его скиллами
2Через Skills CLI (Vercel)
npx skills add EditAtlas/skills --skill video-check -a claude-code3Вручную: положить файлы
~/.claude/skills/video-check/SKILL.mdПапка скилла целиком; для одного проекта — .claude/skills/video-check/
1Через Skills CLI (Vercel)
npx skills add EditAtlas/skills --skill video-check -a codex2Вручную: положить файлы
.agents/skills/video-check/SKILL.mdДля всех проектов — ~/.agents/skills/
1Через Skills CLI (Vercel)
npx skills add EditAtlas/skills --skill video-check -a cursor2Вручную: положить файлы
.cursor/skills/video-check/SKILL.mdCursor читает и .agents/skills/, и .claude/skills/
1GitHub CLI
gh skill install EditAtlas/skills skills/video-check2Вручную: положить файлы
.github/skills/video-check/SKILL.md1Через Skills CLI (Vercel)
npx skills add EditAtlas/skills --skill video-check -a gemini-cli2Командой платформы
gemini skills install https://github.com/EditAtlas/skills --path skills/video-checkВручную: положить файлы
.devin/skills/video-check/SKILL.mdСтарый путь .windsurf/skills/ тоже читается
Claude.ai
Упакуйте папку в zip и загрузите в Settings → Capabilities → Skills
По шагам, с командами для каждой платформы: Как установить скилл в Claude Code: пять способов
Вопросы о video-check
Как установить video-check в Claude Code?
Выполните /plugin marketplace add EditAtlas/skills, затем /plugin install video-check@editatlas-skills.
Что делает video-check?
Проверяет локальный готовый ролик: измеряет параметры файла, ищет дефекты изображения и звука, а затем формирует замечания с таймкодами и способами исправления.
Что нужно для работы video-check?
Нужны Python 3.9+, библиотеки numpy и Pillow, а также FFmpeg с ffprobe. Скилл работает офлайн.
Исходный текст
SKILL.md · Apache-2.0
---
name: video-check
description: Check a finished video before the person sees it or posts it, then fix what is wrong. Measures the file (size, frame rate, pixel format, length), picture faults (black or white frames, frozen stretches, one-frame glitches, stutter, a loop that jumps or repeats a frame), and sound (loudness, clipping, silences, audio that ends early or stops abruptly); makes frame sheets at phone size with the caption-safe area marked; then a reviewer who has not seen your plan looks at the sheets and lists problems with a time and a fix. Fix the worst three, render again, check again, at most three rounds. Use it after rendering any edit, and when someone asks to check a video before sending or posting it, says "kiểm tra video trước khi gửi", "xem lại bản dựng", "video này có lỗi gì không", "review my edit", "QA this render". Works on a local file. Not for breaking down someone else's video (use video-breakdown) or for judging footage the person has not edited.
license: Apache-2.0. LICENSE.txt has the complete terms
compatibility: Requires Python 3.9+ with numpy and Pillow, and FFmpeg with ffprobe. Runs offline. Designed for Claude Code and other agents with a shell that can read images; a reviewer subagent is used when the agent can start one.
metadata:
author: EditAtlas
version: "0.2.5"
status: beta
library: techniques
mcp-tools: search_techniques get_technique get_workflow
---
# Video check: find what is wrong with a render before anyone else does
This is video-check 0.2.5.
The person gets a video that has been checked like a professional editor checks an export, and a short
`REVIEW.md` that says what was measured, what was looked at, what was fixed, and what nobody checked.
Three things are true the whole way through:
- **Measure first, then look.** A model looking at frames misses dead stretches and one-frame glitches; FFmpeg
does not. Run `check.py` before you look, and never skip it because the video "looked fine".
- **The person who made it is the worst reviewer.** You know what each shot was meant to show, so you see the
plan instead of the picture. Have someone without your plan look at the sheets (step 3).
- **You cannot hear.** The script measures loudness, peaks and silences. Whether the music fits, whether a
sound is the right sound and whether speech is clear must be listened to by a person. Say so every time.
## When the inputs are missing
You need the rendered file on this machine. If the person only describes a video, or gives a link to a
platform, ask for the file; do not download from TikTok, YouTube or Instagram. If you made the video, you
also know the plan: size, frame rate, length, loudness target and which still stretches and flashes are on
purpose. If the person brings their own edit, ask what it is for (platform, length) and use the defaults
below for the rest.
## Workflow
1. **Measure.** Pass what was planned, so the script can tell a fault from a choice:
```
python scripts/check.py final.mp4 --out work/check1 --loudness -14
```
When you made the video, or the person told you the plan, add it: for example `--expect 1080x1920 --fps 30
--duration 32 --holds 28.5-32 --flashes 12.4`. Never guess size, frame rate or length from the platform: a
960x540 file nobody asked to be 1080p is not a size fault, so leave `--expect` out when nobody planned it.
`--holds` lists still stretches on purpose (an end card, a title held still); `--flashes` lists planned
one-frame flashes and dips; `--loop` for a piece made to loop; `--audio none` for a silent piece. When someone
speaks over music and you have the two tracks as they sit in the mix (from 0 s, before the final loudness pass),
add `--voice voice.wav --music music.wav`: it measures how far the music sits under the voice while someone
speaks (aim 10 to 16 LU) and flags a bed that is buried or covers the words. Online video
is normalized to about -14 LUFS; use -16 for podcasts and -23 for broadcast. It prints findings as `P0/P1/P2
t=… code: what -> fix` and writes `check.json`, `overview_NN.png` and `findings.png`.
`references/reading-the-checks.md` says what each check catches and what it misses.
2. **Fix the P0s now.** Wrong size, wrong frame rate, wrong length, no audio when there should be, clipping:
these are render mistakes, not taste. Fix the command, render again and rerun step 1 before anyone looks.
For loudness, clipping and true peak, fix the mix in the edit if you have it (lower the music, the loudest
sound). Fix `music_buried` or `music_masks_voice` in the mix too: set the music level under speech once, not a
low gain, then ducking, then a loudness pass on top. If you only have the finished file, run `python scripts/fix_sound.py in.mp4 out.mp4 --loudness -14`:
it limits the peaks, normalizes in two passes and copies the picture untouched, landing within 0.2 LU of the
target and under -1.5 dBTP. Single-pass `loudnorm` misses its peak target and a plain limiter leaves the mix
quiet; do not hand-tune filters round after round.
3. **Look, with fresh eyes.** Open every overview sheet and `findings.png` with the Read tool. Then get a
second look from someone who has not seen your plan:
- If you can start a subagent, give it only the sheets, the platform and one sentence on what the video is
for. Not your plan, not your code, not your intentions. Use the brief in `references/review-guide.md`.
- If you cannot, look yourself, but cover the plan: describe each sheet as a stranger would, one sentence a
frame, before you judge it. Write "one context" in `REVIEW.md`.
Each problem is written as `t=12.4s · where on screen · what is wrong → the exact change`, with P0 (broken or
unreadable), P1 (a viewer will notice) or P2 (polish). No scores out of ten: a score finds nothing to fix.
Check the list in `references/review-guide.md`, which holds the things that make an edit look made by a
template or a machine: slide-like progress rails and step counters, every line entering the same way, text
under the platform buttons, an empty first frame, numbers nobody gave you.
4. **Fix the three worst, then check again.** Take the P0s and the three worst P1s. For each, change the edit
list or the command, not the rendered file. If the fix is a method you do not know, `search_techniques` the
result you want (for example "duck music under speech", "fade audio at a cut") and `get_technique` its prompt.
Render, then rerun step 1 with `--before work/check1/check.json --out work/check2`; it lists what was fixed,
what is still there and what is new, and draws `before_after.png`. Look at the frames you changed.
5. **Stop after three rounds.** If round three still has a P0 or needs big changes, stop and tell the person
what is left and why, instead of looping. Problems that need their decision (a different song, a shot that
does not exist) go to them as questions.
## Deliver
**First, a version check.** If the EditAtlas MCP is connected, call `get_skill` without a name. If it lists video-check above 0.2.5, the first line of your reply says a newer version exists and how to update it (in Claude Code: `/plugin`, then Marketplaces, editatlas-skills, Update marketplace; elsewhere `get_skill` with name video-check gives the step). Without the MCP, skip it.
Write `REVIEW.md` next to the video, in the person's language, with the template in
`references/review-guide.md`:
- the verdict in one line: ready (nothing left), ready with notes (the notes name every finding still there,
and each P1 left needs something you do not have, such as the edit project), or not ready (a P0 is still
there, or a P1 you could fix);
- what the script measured (size, frame rate, length, loudness, true peak, cuts per minute) and the findings
fixed in each round;
- each finding left in place with its times and its pattern as `check.py` wrote them (for example "one every 12
frames from 0.4 s; every 2-6 frames from 11.7 to 14.9 s"), not only a count;
- what the reviewer saw, and whether it was a separate reviewer or one context;
- what still needs a person: listening to the mix and the music choice, spelling and names on screen, lip sync,
colour on a real phone, and anything you could not fix. Keep the sentence that nobody listened to the sound,
in plain words and only in the person's language.
Show the person `before_after.png` from the last round. Never write that the video "has no errors": write what
was checked.
**Last, ask once how it went.** If the EditAtlas MCP is connected, ask the person, in one short message in their
language, about each thing from EditAtlas you used, by name: the techniques, sounds, music, stock clips and this
skill. Send each answer with `send_feedback`, about `this_item`; an answer only about the whole video goes once, as
`whole_edit`. Send only what they said, and nothing if they do not answer. Ask once per edit: if you or another
skill already asked, skip it.
## References
- `references/reading-the-checks.md`: each measurement, its thresholds, what it catches and misses, and the
results on known videos. Read it when a finding looks wrong or before you set `--holds`.
- `references/review-guide.md`: the brief for the reviewer, the list of tells, the problem format and the
`REVIEW.md` template. Read it at step 3.
© EditAtlas · Apache-2.0 · EditAtlas/skills