Communitygithub.com

video-vip-pro

Ein Premium-Videoeditor für einen Aufruf, optimiert für Talking-Head- und Verkaufsvideos in jeder Branche mit individuellen Markenfarben. Transkribiert Rohmaterial, schneidet Fehlaufnahmen aus, fügt rhythmische Zooms, zweistufige Untertitel, Milchglaseffekte und rund 50 VIP-Effekte hinzu und rendert als 9:16/16:9 MP4.

Was ist video-vip-pro?

Die Fähigkeit "video-vip-pro" ermöglicht es einem KI-Agenten, in einem einzigen Durchlauf ein hochwertiges, professionelles Video aus Rohmaterial zu erstellen, das speziell für Talking-Head- und Verkaufsinhalte konzipiert ist und eine breite Palette von Branchen wie Einzelhandel, Immobilien, Wellness, Gastronomie, Bildung, Technologie, Finanzen, Fitness, Reisen, Nachrichten und Vlogs abdeckt. Der Workflow beginnt mit der Analyse des Videos, der Transkription des Tons (unterstützt Vietnamesisch und andere Sprachen) und der intelligenten Auswahl der besten Takes, wobei Fehlstarts und Pausen entfernt werden. Anschließend erstellt das Skill ein dynamisches Storyboard mit automatischen Kamerabewegungen und rhythmischen Zooms mit Bewegungsunschärfe, zweistufigen Untertiteln, Milchglaseffekten und einer Vielzahl von rund 50 VIP-Effekten, darunter Text hinter dem Kopf, Porträtunschärfe, aus dem Rahmen tretende Personen, Countdowns, Stempel, Risse, Schockwellen und Film-Looks. Nach dem ersten Rendern kann der Benutzer das Video durch einfache Aufforderungen oder mit einem lokalen VIP-Editor (der keine Token verbraucht) verfeinern. Die Fähigkeit ist plattformunabhängig und nutzt Python-Skripte für Analysen, Transkriptionen, Zusammenbau und Rendering, was sie zu einem leistungsstarken Werkzeug für Entwickler macht, die hochwertige Videobearbeitung in ihre KI-Workflows integrieren möchten.

Funktioniert mit~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/leminhhieu0707/video-vip-pro/tree/HEAD

In Ihrer bevorzugten KI fragen

Öffnet einen neuen Chat, in dem dieser Agent-Skill bereits geladen ist.

Dokumentation

Video VIP Pro

A raw video goes in. You return a finished, premium edit in one go. After that the user steers with short prompts ("câu 3 đổi hiệu ứng", "màu cam", "bỏ VHS", "chữ to hơn"), so every edit is rebuilt from a small builder script.

Scripts live in scripts/ next to this file. Run them as python -X utf8 <skill>/scripts/<name>.py. First-time setup is setup.ps1 (Windows) or setup.sh; it also runs scripts/get_sfx.py, which downloads the trend SFX pack (about 130 Mixkit sounds in 19 families) to this computer. If a command fails with a missing module or font, run setup.

Files to read

whenread
always, before planningreferences/edit_brain.md (creative pass, 7 edit styles, variety rules, trend sound design) + references/style.md (every effect + props, camera, theme)
picking effects / colour for the businessthe matching row of references/industries.md
a plan field you are unsure ofgrep references/plan.md (full schema)
non-talking-head genres (montage, slideshow, tutorial, podcast)the matching section of references/recipes.md
starting the buildercopy templates/build_vip.py

One-call workflow (do all of it without asking, unless a choice really changes the result)

  1. Brief. Infer these:

    • Industry, from what is said or seen.
    • Theme:
      • the colour the user named, if any;
      • otherwise the brand's colour (logo, sign, uniform);
      • otherwise the industry default.
    • Aspect: 9:16 unless the source is horizontal and meant for YouTube.
    • Length: keep every good sentence.

    Ask only if the user gave several unrelated files with no hint.

  2. Watch. Run analyze.py SRC --every 2 and Read the sheets. Note:

    • where the face is: this becomes FACE;
    • where empty space is;
    • any burned-in text;
    • the shirt colour, because text on the chest must stay readable.
  3. Transcribe. Run transcribe.py SRC --prompt "<brand, product, names>". Then proofread the .transcript.json:

    • fix misheard words, keeping the word count where possible;
    • write prices the Vietnamese way: "8.190.000đ" or "8,19 triệu".
  4. Choose sentences. Write edited/cuts.json with one entry per sentence you keep, in order:

    • drop retakes (keep the later, cleaner take), false starts, "cắt nhé" and dead air;
    • in = first word −0.08 s, out = last word +0.15 s; add "transcripts" (the source .transcript.json files) so aroll.py also writes aroll.transcript.json for captions and the sound brain.

    Then run aroll.py edited/cuts.json -o edited/aroll.mp4 --speed 1.1. It produces a single lip-sync-safe A-roll plus aroll.json, which holds the sentence start times.

  5. Build. Do the creative pass in references/edit_brain.md first. Pick an edit style, 3 motifs and a rhythm map, then write the storyboard table. Never reuse the previous video's sequence. Copy templates/build_vip.py to edited/ and fill it in:

    • set THEME, FACE, CAP_Y;
    • give each sentence a cap(i, top, text);
    • give each sentence a camera move (cam), with HOOK on sentence 0, a SNAP or EMPH on the 2–3 biggest keywords, and alternating PUSH / PUNCH / PULL on the rest;
    • add one VIP effect every 3–5 s, chosen from the industry row and placed on the word that names it (a price triggers CountUp, "3 điều" triggers NumberList, "đau đầu" triggers Cracks);
    • keep the sound brain on ("sound_design": {"style": "trend", "brain": {...}}, see edit_brain.md §4a): effect sounds follow the picture, story sounds follow the words. Then read sound_brain.py's suggestions and cut anything that does not add meaning;
    • previously "sound_design": "trend" meant: every effect gets a layered sound from the trend pack automatically, so add sounds by hand only where they carry the story (riser into a reveal, silence before a punchline);
    • add 2–4 frost scenes per minute;
    • end with a CTA.
  6. Render. Run build_vip.py, then render.py plan_vip.json --draft. Renders over 2 min run in the background. The first render of each motion item takes 3–5 s; person matting takes about 1 s per second of effect.

  7. Check your own output. Run analyze.py <draft> --every 1 and Read every sheet. Look for:

    • text clipped or covering the face;
    • effects that land on the wrong word;
    • unreadable colours for this theme;
    • black or frozen frames;
    • loudness away from −14 LUFS.

    Fix the builder and re-render; the segment and motion caches make this fast. Allow at most 3 rounds per issue.

  8. Final. Run render.py plan_vip.json (no --draft).

  9. Deliver. Send the MP4 (via SendUserFile if available). Then:

    • give a 3–5 line summary: length, theme, the effects used and on which sentence, and anything you guessed;
    • start the VIP Editor in the background, python -X utf8 <skill>/editor/server.py edited/plan_vip.json, so the user can drag, retime, retype and recolour by hand for 0 token.

Mixed voices. If the video mixes on-camera speech with a voice-over, run scripts/voice_match.py so both sound like one voice (see references/edit_brain.md §5). Scan every clip for good on-camera lines (transcribe them all in one pass) before deciding to use only the voice-over.

Refining by prompt (after the first edit)

  • Always edit build_vip.py, never the generated plan. If the user said they changed things in the VIP Editor, re-read plan_vip.json first and edit the plan directly instead, since a rebuild would drop their hand edits. Tell them which approach you are taking.

  • Map requests to edits:

    requestedit
    "đổi màu …"THEME
    "câu N …"the block for sentence N
    "bỏ / thêm hiệu ứng X"delete or add that motion(...) / layer(...) line
    "chữ to / nhỏ"size
    "chữ thấp / cao"CAP_Y or y
    "zoom nhiều / ít"the cam / cycle moves
    "nhạc khác"MUSIC
    "nhanh hơn"aroll.py --speed (then rebuild)
  • Check before the full render: run render.py plan_vip.json --preview A-B on just the changed range, check it, then do the full render once.

Quality rules

  • Captions. One BlueText per phrase, replacing plain subtitles. At most one **keyword** per line. On footage, use tone: "video"; on a light frost, use tone: "light".
  • Clean frame. Besides the captions, keep at most one other element on screen. Never cover the face. In 9:16, keep text out of the top 10 % and the bottom 20 %.
  • Camera. One camera event every 3–6 s. Big moves (SNAP, EMPH) at most once per 10 s.
  • Popout crop. The popout crop must start below the chin, otherwise the head won't break out.
  • Honesty. Use only prices, specs, names and quotes that appear in the footage or the user's message.
  • Licensed media only. Use the bundled synthesized SFX, the user's own files, or royalty-free music. Never use copyrighted songs or reuploaded clips.

Environment notes

  • Python and Node. Python 3.10+ with requirements.txt (faster-whisper, rembg, onnxruntime, opencv, numpy, fonttools, Pillow, edge-tts) and Node 18+ for motion/. Run npm ci inside motion/ (setup does this).
  • FFmpeg. It must be a full build with libass.
  • GPU.
    • An NVIDIA GPU speeds up Whisper and NVENC encoding.
    • Without one, Whisper runs on the CPU; use --model small for speed.
    • rembg uses DirectML or CUDA when available.
  • Fonts. Be Vietnam Pro and Playfair Display are bundled in assets/fonts and motion/public/fonts.
  • Windows printing. Set PYTHONIOENCODING=utf-8 before printing Vietnamese from ad-hoc snippets.
  • Caches. They live in <plan>_work/. Delete that folder only if the cache is broken.

Troubleshooting

  • A motion comp renders blank: render a still with motion/render_motion.py COMP '<props>' --duration 2 --still 1 -o x.png and look at it.
  • Wrong person in the matte (a logo or poster on the wall): use popout "head": [x, w] to limit the column that may break out, or pick a calmer moment.
  • Sound drifts after cuts: never join many talking clips segment by segment. Always go through aroll.py.
  • "sound effect not found": recipe names map to built-in sounds (see SFX_ALIASES in render.py). Drop your own name.mp3 into assets/sfx/ to override one.
  • An effect the scripts can't do: render it with a direct ffmpeg command, or write a new comp in motion/src/blue/ with colours wrapped in T('…'), then use it in the plan.

Verwandte Skills