Video & Animation

Transform a user-supplied photo (semantic reference only) or a text theme into a dense, flat, naïve Outsider Art poster: a map-like mixed-projection world built from large flat color zones, each zone carrying one uniform hand-made texture (window dot-grids, stitch dashes, woven riso grain, ink contour), populated by tiny solid-silhouette faceless figures (never stick figures) doing ordinary things. Every image uses paper white, warm ink black, and a narrow set of 2–4 subdued scene-derived zone c

Split AI content into two layers — AI generates raw material (backgrounds, characters, clips), code assembles the deliverable (text, layout, render). Use when AI-generated carousels/cards come out inconsistent, Thai fonts break, or every re-gen is a gamble.

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.

Honest short-form-video marketing skill (TikTok + Reels/Shorts) for SaaS & apps. Never fabricates market numbers; compliance hard gate; opt-in federated trend dataset.

Cut a long-form recording down to a 60s vertical short. Hook in the first 3s, fast cadence, burned-in captions, 9:16 aspect. Loaded when project_type=shorts or the user asks for short-form output.

Seedance 2.0 prompt engineer. Three modes: WRITE (natural language → production-grade prompt), LINT (diagnose anti-patterns in existing prompt), FIX (auto-rewrite a bad prompt into a good one with diff explanation). Auto-picks T2V/I2V/R2V/V2V/Frames, enforces eight-element formula, three-segment timeline, @-syntax, named cameras, and negative constraints. Output pastes cleanly into ANY Seedance client (jimeng.jianying.com, dreamina.capcut.com, Volcengine API, BytePlus, third-party wrappers). Zer

在 Windows 与剪映专业版中完成默认 9:16、100 秒左右 Vlog 的 L3 规划—拍摄—剪辑闭环。用于把自然语言想法转成低门槛拍摄脚本，按事实时间线和镜头状态匹配冗余素材、补足场景过场、建立连续时间线，并在剪映完成画面、原声、旁白、智能识别人物对白、统一主字幕、少量章节标题、克制的轻网感画面/人物特效和音效，以及固定电视关机＋彩色“晚安”片尾后，通过本地曲库与 FFmpeg 自动或半自动完成选曲、片段切换踩点、人声闪避、混音、审片、成片导出，以及可编辑剪映草稿的本地备份与手机端云同步。以“远景/中景/近景＋画面行为＋可选角度”指导拍摄，不用摄影技术细节拒绝可用素材；只在不可替代故事段完全缺失或真实技术阻塞时暂停。

Short-form content creation: 5 Reels/TikTok/Shorts scripts with hooks, platform-native formatting, and a posting schedule, optimized for each platform's algorithm. TRIGGER: "cria uns reels", "TikTok script", "ideias para shorts", "conteúdo curto", "vertical video", "/content short". SKIP: 10+ minute video or podcast script -> content/script-structure (long-form retention structure); engineering shareability of a piece -> content/viral-design (STEPPS scoring, not production); hooks alone -> conte

创建美观、实用的演示材料。先生成 HTML 交互演示稿，可选导出 PPT/PDF。 **触发场景：** 用户想要制作演示、PPT、分享稿、演讲内容等。 **中文触发词：** "做演示"、"做 PPT"、"做分享"、"演讲稿"、"培训材料"、"答辩"、"路演"、"汇报"、"产品介绍"、" pitch deck"、"演示稿"、"课件"。 **英文触发词：** "presentation"、"slides"、"slide deck"、"pitch deck"、"create slides"、"make a presentation"。 **即使用户只说 "帮我做个分享" 或 "我要汇报一下" 而没有明确说 PPT，也应使用此 skill。**

Idea Darwin Engine — an automated idea iteration system that evolves raw ideas through structured competition and selection. Imports ideas from ideas.md, structures them into scored cards, then runs iterative rounds of deepening, derivation, crossbreeding, critique, and validation to surface high-potential ideas. Use /idea-darwin to trigger. This skill should also trigger whenever the user mentions "run a round", "iterate ideas", "idea pool", "idea scoring", "idea status", "brainstorm iteration"

Teaches AI coding agents how to work effectively with the SMR (Summary-data-based Mendelian Randomization) C++ codebase. Covers architecture, data formats, build system, statistical methods, and common pitfalls.

AI辅助的视频编辑工作流程，用于剪辑、构建和增强实拍素材。涵盖从原始拍摄到FFmpeg、Remotion、ElevenLabs、fal.ai，再到Descript或CapCut最终润色的完整流程。适用于用户想要编辑视频、剪辑素材、制作vlog或构建视频内容的情况。

CapCut (剪映) one-tap video creation on Android. Always clear A_latest first (copy_images_to_album clear-only), then retrieve by time (list_gallery_images) or content (image-memories). Copy to A_latest (clear_album_first), CapCut 照片视频 → A_latest → 全部, multi_tap, N=K.

使用剪映(Jianying/小云雀)的 Seedance 2.0 模型自动生成AI视频。支持文生视频(T2V)、图生视频(I2V)、参考视频生成(V2V)和向后延伸(Extend)四种模式。当用户需要生成AI视频、使用Seedance模型创作短片、基于参考图像/视频进行风格转换，或对已有结果继续延长时使用此技能。需要预先配置 cookies.json 登录凭证。

This skill should be used when creating, improving, or troubleshooting Seedance 2.0 video on any surface - Dreamina, Jimeng, CapCut, Doubao, Volcengine/Ark, BytePlus, Runway's Seedance route, fal, or third-party provider/router surfaces such as EvoLink, OpenRouter, Kie.ai, PiAPI, LaoZhang, Runware, ModelsLab, AI/ML API, MuAPI, SeeGen, and Segmind - including text/image/video/reference-to-video prompts, first/last frame, dialogue, lip-sync and audio, IP-safe rewrites, API, pricing and model-ID qu

AI 短剧 (AI 微短剧) (AI 短剧 (AI 微短剧) (从业者视角) — 用 AIGC 工具（AI 视频生成 / AI 脚本 / AI 生图 / AI 配音 / 对口型 / 角色一致性）制作竖屏微短剧并变现的新兴行业。覆盖: (a) 创作工作流 — 选题→剧本/分镜→AI 生图→图生视频(首尾帧/运镜/参考)→配音/对口型→剪辑/调色→成片，其中角色/场景一致性是第一性难题; (b) 工具栈 — 视频生成(可灵 Kling / 即梦 Seedance / 海螺 Hailuo / Vidu / Runway Gen / OpenAI Sora / Pika / Luma / Pixverse)、脚本(LLM)、生图(Midjourney/SD/即梦)、配音(TTS)、对口型、剪辑(剪映/CapCut); (c) 商业/变现 — 国内付费投流(IAP + ROI + 平台分账, 抖音/快手/红果/微信小程序短剧)与海外订阅(ReelShort/DramaBox/ShortMax/GoodShort)两套打法、品牌定制、成本结构; (d) 合规 — 中国《人工智能生成合成内容标识办法

即梦AI平台(jimeng.jianying.com)浏览器登录。当需要操作即梦数字人、视频生成等功能前检测到未登录时触发。处理协议同意、抖音OAuth扫码、登录态持久化。

Create original short videos using the 界面能力证明蒙太奇｜输入、工具、生成结果与复盘回标 Creative DNA for MiniMax H3 or Seedance 2.0. Use when the user wants this causal, camera, motion, rhythm, or payoff structure with new subjects and surfaces.

AI辅助的视频编辑工作流程，用于剪辑、构建和增强实拍素材。涵盖从原始拍摄到FFmpeg、Remotion、ElevenLabs、fal.ai，再到Descript或CapCut最终润色的完整流程。适用于用户想要编辑视频、剪辑素材、制作vlog或构建视频内容的情况。

How to make an analog-horror PSA short — the stenciled-pictogram / robo-broadcast / VHS-overlay format (faceless, 10 scenes × ~3s, 9:16, ~30s total). A domain overlay on the standard pipeline that supplies the IF / DO-NOT / BUT / AND scenario structure, the locked yellow-on-black 1970s-civil-defense pictogram visual STYLE, the chroma-key-to-alpha rule (icons must be transparent PNGs before composition), the robo-PSA voice profile (ElevenLabs "Alerter" community voice + ALL CAPS input + stability

Tạo video tin tức ngắn 9:16 (~60s) từ URL bài báo hoặc file .txt tiếng Việt. Trigger khi user yêu cầu tạo video tin tức, làm short news, làm bản tin video, render tin thành video, làm TikTok tin tức. Output: video.mp4 + voice.mp3 + script.txt cho CapCut.

The repeatable workflow for short Thai documentary/explainer videos — one topic in, one finished MP4 out. Use when producing a video end to end or running any stage: research a topic, write a script, run an adversarial agent debate over the hook/punchline, generate ElevenLabs voiceover, gather visuals (self-shot, Creative-Commons, or generative), lay timecoded inserts on a timeline, and package a delivery folder. Format-agnostic: aspect/resolution/fps come from a per-project config (9:16, 16:9,

AI-assisted video editing workflows for cutting, structuring, and augmenting real footage. Covers the full pipeline from raw capture through FFmpeg, Remotion, ElevenLabs, fal.ai, and final polish in Descript or CapCut. Use when the user wants to edit video, cut footage, create vlogs, or build video content.

Use this skill whenever the user wants to download media (videos, images, audio) from social media platforms using the btch-downloader npm client SDK. Triggers include: downloading from Instagram, TikTok, YouTube, Facebook, Twitter/X, Pinterest, Spotify, SoundCloud, Threads, CapCut, Google Drive, MediaFire, Douyin, SnackVideo, Xiaohongshu, Cocofun, or Kuaishou using JavaScript/TypeScript; using the igdl, ttdl, fbdown, youtube, twitter, or aio functions; building a media downloader bot or app in