Video & Animation

Translate, repair, or selectively polish English-to-Hebrew SRT subtitles. Use for Hebrew subtitle translation, synchronized SRT repair, RTL punctuation, SDH cleanup, character-aware gender, recurring series terminology, legacy Hebrew encodings, translator/site credit removal, full bilingual editorial review, and batch or ZIP subtitle QA.

Replace a video's original English vocals with Chinese voiceover on an identity timeline — the video keeps its original speed and duration, each cue's Chinese audio is fitted into its original window. Use when the user wants to dub a video into Chinese — mentions 中配 / 配音 / 中文配音 / 换原声, or has a cooked bilingual video and wants a second Chinese-narrated release, or another skill (e.g. video-cooking) hands off "video is done with subtitles, add a Chinese dub.

Automated video localization pipeline for YouTube, X/Twitter, Instagram, TikTok and Bilibili. Downloads a video, produces bilingual or Chinese-only subtitles, and burns a ready-to-publish release video. Use whenever the user asks to download a video, generate or translate subtitles, transcribe audio, make a Chinese-subtitled (熟肉) version, or localize streaming media. Requires Python >=3.11 and FFmpeg with libass. Transcription needs an OpenAI-compatible Whisper API key or the VideoCaptioner CLI; translation needs no key.

翻译 YouTube 录播视频字幕，生成浏览器插件可导入的 JSON 和译文 SRT；支持首次工具目录设置、继续翻译、术语返修、迁移目录和更换本地识别模型。Translate YouTube subtitles with the current session model. Use for video subtitle translation and managing this subtitle tool, not generic text translation or live video.

Design, implement, render, review or improve animated S03–S12 scenes in arranque-so, or remake S01/S02, with characters, cameras, signals, a 3D world and dialogue synchronization. Use when working on cinematic choreography and validating technical fidelity and continuity.

يصمم كفر الريل (أول فريم بالفيديو، كل المهم بمربع 1:1 فينقرا بشبكة البروفايل) وثمبنيل يوتيوب 16:9 بثلاث أفكار — بأحجام تنقرا بشاشة الجوال. Use when user says "سو كفر"، "كفر الريل"، "غلاف الريل"، "ثمبنيل"، "ثمب يوتيوب"، "صورة مصغرة"، "thumbnail"، "الكفر ما يبين"، or after any finished video. Not for carousel covers. Darija triggers: "دير لي غلاف"، "دير thumbnail".

مونتاج مقاطع بلا كلام على الإيقاع — مجلد مقاطع (مقهى، سفرة، منتج، مكان، فعالية) يتحوّل لفيديو واحد 9:16 يختار أحلى لحظة من كل مقطع ويقطع على نبضات الصوت. Use when user says "مونتاج من هالمقاطع"، "سو فيديو من هالكلبات"، "ركّب المقاطع على الأغنية"، "مونتاج على الإيقاع"، or sends a folder of clips without speech. Not for a talking video (use dz-reel). Darija triggers: "دير لي مونتاج على الإيقاع"، "ركّب هاد المقاطع".

فيلم موشن قرافيكس ثلاثي الأبعاد بلا شخص ولا تصوير — من نص أو فكرة أو سكربت، بصوته أو صوت مولّد، 60 فريم. وشرح موقع أو تطبيق بلقطات حقيقية داخل آيفون أو ماك بوك 3D. Use when user says "فيديو موشن"، "شرح بالموشن"، "سو فيديو من هالنص"، "موشن قرافيكس"، "اشرح هالموقع بفيديو"، "اشرح التطبيق بفيديو"، "ديمو آيفون". Not for editing a talking video (use dz-reel). Darija triggers: "دير لي فيديو موشن"، "اشرح بالموشن".

مونتاج فيديو هدرة للكاميرا (talking head) إلى ريل 9:16 بالدارجة الجزائرية — ينحّي السكتات والغلطات، كابشن عربي كلمة بكلمة مع ترجمة عربية للكلمات الفرنسية/الإنجليزية، موشن قرافيك، مؤثرات صوتية وغلاف. Use when the user says "مونتيلي هاد الفيديو"، "دير لي ريل"، "نحّي السكتات"، "دير كابشن بالعربية"، "monter cette vidéo"، "edit this video into a reel", or sends a talking-head video. Not for carousels or video covers (use dz-thumbnail).

يمنتج البودكاست والمقابلات — كاميرتين أو أكثر لنفس الجلسة إلى ريلات 9:16 بكابشن عربي وهوك مكتوب، أو الحلقة كاملة 16:9 لليوتيوب، أو حلقة ساعة إلى عدة ريلات. Use when user says "بودكاست"، "كاميرتين"، "الحلقة كاملة بالعرض"، "حلقة يوتيوب"، "ريلات من الحلقة"، "طلّع ريلات من المقابلة"، or sends multi-camera files or a long episode. Not for a single talking video (use dz-reel). Darija triggers: "دير لي ريلز من البودكاست"، "قطّع الحلقة".

Dựng video người nói dọc 9:16 theo phong cách Hanta AI, với chữ vàng Anton/Montserrat, thẻ đồ họa, cảnh chèn không che người và CTA; chép lời, lập edit.json theo mốc nguồn và xuất MP4 bằng Remotion. Dùng cho talking-head hoặc video chuyên gia theo mẫu này, không dùng cho montage tour có lời thuyết minh.

Dựng video quảng bá tour ngang 16:9 có lời thuyết minh theo lịch trình ngày, nhãn địa danh, phụ đề, sơ đồ hành trình/tàu, logo và CTA bằng Remotion; lấy phong cách từ Cửu Trại Câu V6 giọng Thục Đoan. Dùng khi có tour hoặc lịch trình cần dựng, không dùng cho talking-head hay video chỉ cắt theo nhạc.

Create directed product motion graphics, kinetic typography, interactive HTML animation prototypes, iOS-style wheel pickers and rolling counters, deterministic HyperFrames films, 3D device launch promos, and locally composed MIDI scores. Use for launch videos, UI showcases, animated controls, prompt-bar camera moves, HTML-to-video workflows, and music timed to product actions.

Author, animate, and render high-fidelity EdTech and Moodle LMS video compositions using HyperFrames. Covers step-by-step tutorials, course showcases, tripartite structures, SCORM gamification demos, and responsive 16:9 / 9:16 video generation.

Draft captions from source-bound transcripts onto an edited Cutbolt timeline and attach editable scene clips without an intermediate movie.

Edit a talking head clip into a finished 9:16 Instagram reel with real captures, logos, motion graphics, captions, music and sound. Use when asked to edit, cut, make or improve a reel or short from a video file, or to apply a note to a reel draft.

Make beat-synced vertical (or 16:9) promo reels from code with the Kinetix pipeline — scenes as HTML/JS, a MusicGen or pure-synth soundtrack with drops placed by beat, word-timed captions, Remotion light leaks and stingers, a HyperFrames quality gate, and a render with a baked-in thumbnail. Use when someone asks for a reel, Short, TikTok, hype/launch/demo video, "make a video about my app/project", or wants to edit/re-cut one made with Kinetix.

Transforma um vídeo longo (transcrição + metadados) em cortes verticais (Shorts/TikTok/Reels) otimizados para a linha de retenção, usando a estrutura Hook → Storytelling → Payoff. Seleciona os trechos, pontua, reordena o gancho, define cortes, legendas, zooms, música e efeitos, e devolve uma Edit Decision List (EDL) em JSON.

Make a narrated infographic explainer film (typically 5–15 minutes) about a real subject — how a system works, what happened, what happens next — as a 1950s instructional-card newsreel with a pantomime mascot, an over-eager narrator, sung commercial breaks and code-drawn infographics that land on the spoken word. Pipeline: research → fact-checked script → ElevenLabs narrator with per-character timestamps → Suno theme, beds and jingles → Gemini illustrations → Remotion scenes cued to words → −14 LUFS master + SRT. Use when the user asks for an explainer film, infographic film/video, "a film that explains …", "gör en film om …", a newsreel-style video about a topic, or another film "like the election film".

Revise or write technical and scientific prose so it reads as precise academic writing, not AI-generated or colloquial text. Covers the recurring tells (everyday verbs for mathematical operations, objects given agency, pictures instead of guarantees, paraphrases of named concepts, slogans as reasons, spatial metaphors, fillers), what to leave alone, a marked-edit workflow that keeps numbers, citations and meaning, and paper-level narrative (launch principle, 发布会原则: lead with the strongest advantage, no self-weakening wording). Use whenever the user asks to write or restructure an abstract, introduction or conclusion, frame results against baselines, polish, restyle, tighten, proofread, "make it sound academic", "remove the AI tone", or run a style pass on a paper, SI, caption, rebuttal or report; when a co-author or reviewer calls the text colloquial or AI-like; and before handing over any new paper prose you wrote, even unasked.

Transforma um vídeo longo (transcrição + metadados) em cortes verticais (Shorts/TikTok/Reels) otimizados para a linha de retenção, usando a estrutura Hook → Storytelling → Payoff. Seleciona os trechos, pontua, reordena o gancho, define cortes, legendas, zooms, música e efeitos, e devolve uma Edit Decision List (EDL) em JSON.

Make a smooth, Screen Studio-style demo video of a web app from real footage — scripted headless 60 fps capture, spring auto-zoom camera, re-rendered cursor, padded stage, subtitles and callouts — with the webapp-screencast CLI. Use when asked to record a product demo / walkthrough / launch video / screen recording of a web app, to "make it look like Screen Studio", or when /brag (or any launch-video skill) should use real app footage instead of screenshots.

Short technical films made by any coding agent, drawn and scored entirely in code (HyperFrames) from a mode template. Ten modes — Thread, Ink, Plakat, Blueprint, Phosphor, Tally, Trace, Morph, Halftone, Pulse — each with palette presets, a form variant, a scene registry and a renderable template; length follows the content, pace is brisk, every render is measured. Use when asked to make, remake or restyle a launch, promo, teaser, showreel, data story or explainer video for a technical product, or to choose a visual style for one.

This skill should be used when the user asks for the finished video file - "exportier das", "render das fertig", "mach das final", "renderte das Video", "mach den Endexport", "bau das Video zusammen", "mach es fertig für YouTube", "gib mir die MP4", "export the final", "render the final cut", "make the master", "export for TikTok", "give me the mp4". Covers segment rendering with the cache, guarded concat, the one-pass audio, loudness normalisation and the single mux.