Video & Animation

Use when the user asks for an animated product explainer, launch video, demo reel or social video in this house style (dark navy stage, stylized UI cards, a cartoon cursor that glides and clicks, captions at the bottom), rendered from a single HTML file to MP4 with an optional synthesized music bed. Also use when editing or re-rendering an existing video built this way.

>- Post-mortem on what the user has already posted - which reels actually worked, why, and what to stop making. Use when the user pastes their Instagram insights or past posts and asks "what's working", "why did this flop", "read my analytics", "audit my content", or wants to know what to do more of.

>- Write the Instagram caption - the line that survives the "... more" cut, the body, the single ask, the search terms and the three hashtags - and lint it before it goes out. Use when the user says "write the caption", "caption this", "what do I put in the description", has a reel or a carousel ready and needs the text, or asks about hashtags.

>- Write comments on other people's Instagram posts and reels that read as a person with an opinion, not a bot. Use when the user pastes a post or a reel and wants a comment, says "comment on this", "engage with this", "what do I say here", or wants a batch for their daily engagement round.

Use when making or revising any motion graphics piece (launch film, explainer animation, YouTube insert, whiteboard graphic, social/vertical clip, showreel, one-shot showcase) and before writing scene code; also when a motion piece looks like slides, a website or generic AI output, or feels static or slow. Trigger words: Motion Graphics, Animation, Launchvideo, Erklärvideo, Einblendung, Showcase, Stil, \"wie ein Motion Designer\".

Use when a motion graphics piece needs references or inspiration (launch videos, styles, camera moves), when a reference clip should be downloaded, studied or measured, or when a render is to be compared against a reference or checked for slide/slop patterns. Trigger words: Inspo, Referenz, whatships, Launchvideo-Beispiele, Stil finden, ref_metrics, motion_check, Kontaktbogen.

Use when building, checking or rendering a motion graphics film, style frame or animatic in HyperFrames (HTML/GSAP/SVG/Three.js compositions), when setting up a HyperFrames project, or when a HyperFrames render shows banding, wrong colours, black backgrounds, missing frames or missing motion blur.

Use when a motion graphics piece needs real 3D (product shot, metal, glass, depth of field, macro, object that transforms or assembles, camera shot list) in HyperFrames via Three.js, or when 3D models should come from Blender (Blender MCP, headless Blender, glTF/GLB export). Trigger words: 3D, Produktfilm, Metall, Glas, Tiefenunschärfe, Bokeh, Three.js, Blender, GLB.

Use when a motion graphics film, animation or video insert needs sound (music choice, sound effects, cue list, ducking, loudness master) or when sounds are added to the local sound library. Trigger words: Sound, Ton, Musik, SFX, Geräusch, Whoosh, Cue-Liste, Mischung, LUFS.

Make a short vertical explainer reel (Instagram Reels / TikTok / Shorts, 9:16) in the "Pixel Crew" style — pixel-robot AI employees, chunky offset-shadow cards, word-synced captions, voiceover, chiptune bed and SFX — rendered locally to MP4. Use when the user asks for a reel, short, TikTok or viral video about AI agents, automations, a team/workflow, a list of tools or steps, or wants "a reel like the delivery-team one".

给一个 GitHub / 开源项目链接，产出两条成片——约 30 秒的商用级宣传片（promo）和 1–3 分钟的精准介绍片（explainer）。HTML+GSAP 确定性时间轴逐帧渲染、MiniMax 中文配音、代码合成配乐音效、独立审查 agent 打分返工。Use when the user gives a repo/open-source link and asks for a promo, trailer, launch video, 宣传片, 介绍视频, 讲解视频.

Analyze reference Reels or videos for editing rhythm and visual effects, then document reusable, evidence-backed techniques in the local effect library. Use for video-style analysis, not for generating replacement footage.

Look up video-editing effects, transitions and looks (whip pan, speed ramp, match cut, VHS, word-by-word captions, split screen, rack focus and about 120 more) with practical recipes before building them in an edit. Use whenever the user asks for an effect, a transition, a look or "something like in that reel", or when you plan transitions, emphasis or captions for a cut.

Biên tập VLOG / video nói chuyện quay thô (talking-head, quay trên xe, chia sẻ ngoài trời, 16:9) thành video hoàn chỉnh bằng HyperFrames: bóc lời Whisper theo từng chữ, AI chọn đoạn trọng điểm theo thông điệp và khán giả rồi cắt khớp khung (không cụt chữ), lọc ồn và chuẩn âm −15 LUFS, chèn cảnh trám tắt tiếng 4–5 giây phủ lên lời nói, punch-in che jump cut, chỉnh màu, chữ động tiếng Việt (tiêu đề, chương, câu chốt, con số, danh sách, sơ đồ quy trình, lời mời). Dựa trên Rule of Six của Walter Murch và cách dựng vlog của Casey Neistat. Dùng khi người dùng muốn rút ngắn video dài, lấy trọng điểm, dựng vlog, thêm cảnh trám/b-roll, lọc âm, chỉnh màu, chèn chữ cho video quay bằng DJI/Osmo/điện thoại. Kích hoạt khi có từ: biên tập video, dựng vlog, edit vlog, rút ngắn video, lấy trọng điểm, cắt video, chèn cảnh trám, b-roll, lọc âm, chỉnh màu video, chèn text vào video, video lái xe chia sẻ, talking head, video DJI, Osmo Action, hyperframes vlog.

The director of a reel made from the creator's own footage — runs the whole job from idea to a post-ready MP4 + cover + caption and makes the creative calls (which story structure, which 3–5 effects go on which beat, where the transitions are, pacing, sound), then reviews the cut like a critic before handing it over. Use when someone says "make me a reel", "turn my clips into a reel", "edit this like a pro", "what do you think as a critic / director", "make it more alive", "add effects", "зроби рілс", "що скажеш як критик", or when a reel is being planned, reviewed or re-cut.

Create videos with interchangeable video-model APIs, mandatory Manim rendering, Kokoro Chinese narration, subtitles, provider switching, cost safeguards, bounded retries, and usage accounting. Use across Codex, Claude, Gemini, OpenClaw, Hermes, DeepSeek, Kimi, WorkBuddy, or normal Python scripts.

Write, analyse and plan short-video scripts (Instagram Reels, TikTok, YouTube Shorts, 30–60 s) in four proven viral structures — contrarian insight with a named archetype, series episode with a cliffhanger, humor hook + list of N factors, and "the scenario of …" confession. Use when the user asks to "write a reel script", "give me reel ideas", "analyse this viral reel", "break this reel down", "turn this transcript into a structure", "make one like this reel", pastes a reel transcript, or wants to edit a reel that has no script yet. Triggers in any language (e.g. "напиши сценарій для рілса"). Writes flowing spoken text in the creator's own language (English, Ukrainian, …), checks it against a checklist, and hands a montage plan to reel-builder, talking-head-edit or reel-director.

Build a finished voice-over reel (Instagram Reels / TikTok / Shorts, 1080×1920, 25–70 s) from one reel.json — the creator's own clips and photos + a voice-over + subtitles with accent words + hook/end cards, stamps, punch-ins, transitions and motion effects (cover story, orbit text, doodle boil, glass type, checklist, headline wall, black-and-white → colour burst, photo montage), music bed, −14 LUFS mix, safe-zone QA sheet, cover and caption. Use after a script is approved (reel-script skill) when the reel is a voice-over over footage (for someone talking to camera use talking-head-edit), or when asked to "make / build / rebuild / re-cut a reel", "змонтуй рілс", "перезбери".

Operate the local VideoHero app through its supported CLI to discover voices and models, generate speech from text, render PPTX/PDF presentations as narrated videos, and monitor or cancel VideoHero jobs. Use when a user asks an agent to create media with VideoHero; do not use for generic media tools or UI-click automation.

Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated motion-first unit, including an animated title. Route fresh creation through hyperframes before using this skill.

Import Figma content into a HyperFrames composition — rendered assets, brand tokens, components, storyboard sections → reconstructed motion (frames read as states, not slides) (REST/CLI), connector-assisted motion when available, and shaders from a connector or native export. Use when the user pastes a figma.com link or asks to bring a Figma design, frame, logo, brand, or animation into a video/composition.

Turn arbitrary text — an article, notes, a topic, a brief — into a faceless explainer video: there is no site or footage to capture, so the visuals are invented per scene (typography, abstract graphics, diagrams, data-viz). Use for topic explainers, concept breakdowns, how-tos, listicles. Not a video built from a website (/product-launch-video — promo or tour). Unclear → /hyperframes.

Research Instagram Reels with the reels_* MCP tools: read Reels from their URLs (details, timestamped transcript, real frames), extract hooks, quotes and on-screen text across many Reels, compare how creators open their Reels, and collect Reel links from web pages. Use for requests like 'analyze the hooks of these Reels', 'what do these creators say about X', 'find Reels about Y and break down how they open'.

Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet `anchor` rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.