All authors
Claude Skills by Wanfeng1028
github.com/Wanfeng1028590 skills0 installs0 views
- Kunka Screenwriting Master Kunka Scene创作或改写单场戏、场景衔接和动作描写，调整进入时机、信息揭示与局部节奏。 Write or revise individual dramatic scenes.Votes: 0GitHub stars: 5
- Kunka Screenwriting Master Kunka Screenwriter昆卡编剧大师总入口。启动或续写剧本项目、把零散想法推进为可写大纲，或用户不确定下一步时使用。单场戏、对白润色与局部诊断直接交给对应技能。 Start or resume a screenplay project.Votes: 0GitHub stars: 5
- Kunka Screenwriting Master Kunka Series开发短剧、连续剧、单元剧的分集大纲、季线和角色跨集变化。处理本集兑现与后续悬念的平衡。 Plan episodic drama and series arcs.Votes: 0GitHub stars: 5
- Kunka Screenwriting Master Kunka Structure编排剧本大纲、转折、叙事顺序与多线结构，诊断因果和节奏断点。用于故事层面的组织问题。 Outline and diagnose screenplay structure.Votes: 0GitHub stars: 5
- Kunka Screenwriting Master Kunka Style Lab用同一戏剧情境比较观察式、群像、悬疑或喜剧等叙事手法，探索剧本语气与节奏。适用于风格实验与版本比较。 Explore dramatic styles through original scenes.Votes: 0GitHub stars: 5
- Lanshu Awesome Ai Video Kit Happyhorse Prompter生成符合 HappyHorse 1.0 严格规则的紧凑提示词（30-55 词），主体先行 + 明确镜头技术 + 音频激活路径（with X audible / speaking English at natural pace）。可选加入"8s 时序节拍"结构。用于"用 HappyHorse 生成视频"、"做个 3-15 秒短片"、"要原生带音频的视频"、"ASMR 视频提示词"等触发场景。Votes: 0GitHub stars: 5
- Lanshu Awesome Ai Video Kit Model Selector根据用户的视频需求(场景、时长、音频、语言、平台限制、预算、是否需要本地部署/角色一致性等),从 16 个主流 AI 视频模型(12 商业 + 4 开源)中推荐最匹配的 1-3 个,并解释为什么。覆盖商业:Seedance 2.0、HappyHorse 1.0、Kling 3.0、Sora 2、Veo 3.1、**Gemini Omni**(2026-05 新)、Runway Gen-4.5/Aleph、Pika 2.5、Hailuo 02、Hunyuan Video 1.5、Wan 2.7、即梦 AI;开源:LTX-Video 0.9.7、Mochi 1、CogVideoX 5B、Higgsfield Soul。用于"用哪个 AI 视频模型好"、"Sora 还是 Kling"、"国内有什么模型"、"哪个免费"、"哪个支持中文"、"哪个能编辑已有视频"、"哪个能本地部署"、"哪个角色一致性最强"等触发场景。Votes: 0GitHub stars: 5
- Lanshu Awesome Ai Video Kit Prompt Translator把一条 AI 视频提示词从源模型(如 Sora 2)的写法风格转换为目标模型(如 Kling 3.0 / Wan 2.7 / Veo 3.1 等)的最佳实践写法。基于 110 条 10 场景 × 11 模型对照基准数据(prompts/data/cross-model-matrix.json),不是凭直觉重写,而是查表式 in-context learning。10 场景:产品 / 双人对话 / 物理动作 / 图生视频 / 多人会议 / 恐怖 / 自然延时 / 抽象 / 武侠 / 萌宠。用于"Sora 已 EOL 帮我把这条提示词改成 Veo"、"我有 Kling 提示词想跑 Wan"、"跨模型 A/B 测试"、"把英文 prompt 优化成 Kling 中文版"等触发场景。Votes: 0GitHub stars: 5
- Lanshu Awesome Ai Video Kit Seedance Debugger诊断 Seedance 2.0 生成视频时出现的常见问题（人物 ID 漂移/双胞胎/字幕/Logo/风格漂移/延长跳变/画质劣化/特效不对/中文发音/音色不准/结尾噪音 等 12 类），定位根因并给出修复后的提示词。用于"我的提示词生成出来不对"、"视频里出现奇怪的字幕"、"人脸不像参考图"、"出现两个一样的人物"、"风格变了"、"怎么修这个提示词"等触发场景。Votes: 0GitHub stars: 5
- Lanshu Awesome Ai Video Kit Seedance Prompter把用户的自然语言视频需求转换为符合 Doubao Seedance 2.0 进阶公式的提示词（8 要素：精准主体+动作细节+场景环境+光影色调+镜头运镜+视觉风格+画质+约束条件）。用于"帮我写一个 Seedance 提示词"、"生成视频提示词"、"做个产品广告视频"、"用 Seedance 生成 XX"等触发场景。如果用户没明确说 Seedance，但描述的是单一镜头的复杂叙事/多主体/电影感视频，也优先用此 skill。Votes: 0GitHub stars: 5
- Lanshu Awesome Ai Video Kit Seedance Storyboard把复杂剧情/故事大纲拆分为 Seedance 2.0 的"镜头1/镜头2/镜头3"多分镜结构。每个分镜按 4 维度组织（运镜 + 主体动作与表情 + 位置/空间变化 + 音频）。用于"帮我把这个故事写成 Seedance 提示词"、"分镜脚本"、"多镜头视频"、"剧情复杂的视频"、"剧本转分镜"等触发场景。Votes: 0GitHub stars: 5
- Legal Professional ServicesLegal and professional-services video: attorney introductions, practice-area explainers, FAQ series, case-result and trust films. Formal, restrained, plain-language, with mandatory compliance and advertising-rule handling and an asset-request protocol. Use for law firms, accountants, consultants and advisors.Votes: 0GitHub stars: 5
- Lemo OpuscarDirect and produce a short film made entirely in code, in one of the styles of the Lemo-Opuscar library (e.g. Impasto Oil Painting 油画厚涂, Watercolor Brush 水彩笔刷, Ukiyo-e 浮世绘, Satellite Briefing 卫星简报, Exploded Teardown 产品拆解, Microscope Story 显微叙事). Use when the user asks for a video, film, short, promo, explainer or animation (视频、短片、动画、宣传片) in a named style, including data, science and product explainers built from real material such as satellite imagery, laser scans, CAD files or archive scans ...Votes: 0GitHub stars: 5
- Lingcut Video端到端 AI 影视/短剧流水线。用 LingCut（lingcut.cn）无限画布生成分镜图与视频、LibTV 生成配音与旁白、ffmpeg 拼接成片，并交付一个可直接剪辑的工作台。当用户要求"做一部 AI 短剧/AI 电影/AI 视频/MV"、"用 lingcut 生成视频"、"把剧本做成视频"、"这个剧本拍出来"、"给这段文字配画面"时使用。Votes: 0GitHub stars: 5
- Live Action Drama 16x9将剧本转为16:9真人质感AI短剧的可执行分镜、运镜、光影、人物状态与细微表演设计，输出4—15秒整数生成单元、首帧及视频提示词，并进行连续性与成片质量审查。适用于剧本分镜、真人短剧提示词和已有分镜或视频的审查修订。Votes: 0GitHub stars: 5
- Long Video To Shorts ConfigInteractive configurator for the long-video-to-shorts skill. Use when the user types /long-video-to-shorts-config or asks to "configure/change the shorts style", "change the caption style / accent colour / font / hook duration", "règle le style des shorts", "change la couleur des sous-titres". Walks through the options (keep-or-change each), then writes config.json for them.Votes: 0GitHub stars: 5
- Long Video To ShortsTurn ONE long video into several vertical (9:16) captioned shorts, fully local and free — a home-made Opus Clip. Use when the user wants to "faire des shorts / des clips à partir d''une longue vidéo", "découper une vidéo en clips verticaux", "reels/shorts/tiktok from a video", "extraire les meilleurs moments", or names Opus Clip / Klap / Vizard and wants it done locally. Handles two content types: face-cam (smart-crop to face + embedded-captions) and screen/gameplay (blurred-pad + burned kara...Votes: 0GitHub stars: 5
- Make Prompt Seedance2 Seedance Video Prompt生成 AI 短视频（Seedance 2.0 / 即梦 / 小云雀）分镜提示词的专家。当用户想生成任何形式的短视频提示词、视频脚本、分镜脚本时使用——包括但不限于：剧情/叙事视频、打斗/动作视频、风景旅拍、短剧、伪纪录片、家装空间漫游、UI 产品宣传片、口播视频、音乐卡点、品牌广告片 TVC、产品展示、电商带货视频（抖音/TikTok）、真人感种草/探店视频、视频延长或改剧情。只要用户提到"生成视频提示词"、"帮我写分镜"、"做个短视频"、"带货视频"、"产品广告"等就应触发。Votes: 0GitHub stars: 5
- Makevoid Motion Graphics Music VideoCreate and revise animated character music videos from a supplied song and creative prompt, with researched storyboards, Fal character and video generation, p5 motion graphics, audio editing, and Swift VFX. Use for full music videos or revisions to their characters, scenes, animation, sound, and effects.Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Asset Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】资产锁定阶段专精引擎：从剧本提取三清单（角色外观/道具/场景）→**按素材形态产出**（M0 外观档案复述块 / **M1 出图提示词 + 图号编号 + 《素材接线单》·默认** / M2 分批出图锁定），并管理场景空间连续性、道具单实例锁与人群调度。**本包不持有图**——图由用户在外部平台自行生成；本引擎禁止索要图片、禁止因缺图阻断下游。触发词：拆资产、资产提取、资产图册、生图提示词、角色资产板、场景提示词、道具提示词、三清单、九宫格场景图、人物立绘、场景连续性、道具穿帮、人群调度、素材形态、接线单、我没有图。本技能为套装成员，完整全流程从总控 short-drama-director 进入。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Combat Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】打戏分镜专精引擎——负责一切打斗/武打/战斗场面的分镜提示词生成：打斗构思（动机/强弱错位/赌注/环境变数）、八步写作脑、八类自然句母版逐镜成句、铁律自检、摄影机行为锁与战场群战速查。触发词：打戏、武打、分镜打斗、战斗场面、大招、剑阵、法阵、法相、割草、对轰、招式、武学、拳法、刀剑对决、终结技、战场、群战、军阵、运镜锁。套装成员：总控 short-drama-director（锁系统与投喂格式权威）+ sd-decision-engine（决策）+ sd-script-engine（剧本）+ sd-asset-engine（资产）+ sd-storyboard-engine（分镜）+ sd-combat-engine（打戏·本技能）+ sd-feed-engine（投喂）+ sd-qc-engine（质检）。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Decision Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】导演决策引擎（分镜选型依据层）：情境映射 E01~E15 + 80/15/5 权重机制 + 情绪锚点表 + 构图雷区 A/B 双态白名单，回答'这场戏该用什么镜头'，为分镜引擎提供选型依据。触发词：情境映射、这场戏用什么机位、情绪锚点、节拍表、雷区、构图雷区、正面居中能不能用、这镜该用什么。本技能为套装成员，完整全流程从总控 short-drama-director 进入。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Feed Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】投喂适配引擎（交付层）：分镜稿→可投喂提示词的最后一关。A/B/C 三种输出模式（默认 C 导演轨纯净版）+ 输出前六道闸门（节奏/台词时长/字数/可投喂语义/图片数量/编号一致性**+素材形态一致性**）+ 素材映射与人物锚定 + **《素材接线单》** + 多段衔接策略 + 工程语言清洗。**本包不持有图**——图由用户在外部平台自行生成；默认 M1（编号占位），禁止索要图片、禁止因缺图阻断。触发词：转Seedance、生成提示词、投喂、提示词包、@图N、素材映射、人物锚定、多段衔接、字数超了、清洗工程语言、素材形态、接线单。本技能为套装成员，完整全流程从总控 short-drama-director 进入。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Jieshuo Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能 · 赛道③ 解说叙事类】解说/叙事类视频专精引擎：历史人物志 / 历史事件战役 / 科普知识解说 三类赛道的完整规则集。核心能力——旁白稿（VO Script）写作规范（解说语速 4.5~6.5 字/秒 / 短句 ≤25 字 / 类比优先 / 减负原则 / 钩子与金句）、声画信息差纪律、时代场景优先的历史视觉规范、三大子类型爆款公式、旁白句→画面的分镜适配、C14 画外声配额豁免。触发词：解说、拉片、盘点、历史人物、历史事件、科普、冷知识、知识区、一口气看懂、纪录片解说、口播。本技能为套装成员，完整全流程从总控 short-drama-director 进入；赛道由总控〇·0 分流锁选定。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Overseas Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能 · 赛道② 海外剧】海外剧剧本专精引擎：非中文正文 / 北美等真实地域口语 / 多语 / 方言项目的剧本层规则集。核心能力——语言与风格合同（写作前建立、跨阶段传递，非成稿后润色标签）、三级校准状态（region_unvalidated / evidence_calibrated / human_reviewed，模型不得自行升级）、北美英语对白适配、不从身份推权限、单语听者边界、地方细节准入。触发词：海外剧、英文剧本、北美英语、美式对白、多语、方言、本土化、换语、localization、ESL、字幕。本技能为套装成员，完整全流程从总控 short-drama-director 进入；赛道由总控〇·0 分流锁选定。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Qc Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】成片质检与连续性管理专精引擎：跨镜/跨集一致性台账、剪辑转场设计与 Seedance 15 秒分段门、声音设计模板、后期音乐与声音设计（BGM Cue Sheet/音乐搜索词/AI 音乐提示词/版权检查）、出稿机检使用。**一致性门按素材形态改判**（M0/M1 比文字档案；M2 才做图像级核验；本包不持有图，禁止索要图片、禁止虚假核验断言）。触发词：审查、质检、连续性检查、看片、拉片、台词诊断后复核、剪辑转场、分段检查、声音设计、机检、配BGM、后期音乐、找音乐、Suno提示词、音乐版权、素材形态。本技能为套装成员，完整全流程从总控 short-drama-director 进入。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Real Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】真人写实打斗专精引擎——负责现代格斗/警匪追逐/悬疑暗杀/喜剧打斗/儿童向打斗，以及小规模多人战斗（1v2/1v3/2v2/3v3）的动作设计与 Seedance 2.5/2.0 投喂提示词生产：R1 写实克制档、11 环动作动力链、真实格斗招式库（19 门武学/36 路腿法/擒拿 36 式）、防守三态与闪避大幅化、多人战斗调度、真实受力反馈与合规转译。触发词：真人打斗、写实格斗、现代搏击、警匪追逐、悬疑暗杀、喜剧打斗、MMA、群战、一打多、以少打多。套装成员：总控 short-drama-director（锁系统与投喂格式权威）+ sd-decision-engine（决策）+ sd-script-engine（剧本）+ sd-asset-engine（资产）+ sd-storyboard-engine（文戏分镜）+ sd-combat-engine（通用打戏）+ sd-xuanhuan-engi...Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Recon Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能 · 〇·1 素材来源锁 · 对标采风】素材采风与对标情报引擎（轻量版）：用户说「我要对标某个爆款」时，把该剧的公开信息从全网（含外网）找回来，按固定格式交付一份《对标情报卡》。核心能力——短剧信源分级清单（平台官方榜 / 数据指数站 / 行业媒体 / 社媒讨论 / 外网指数）、采风四步（定目标→定信源→检索→落卡）、对标情报卡十三项字段模板（含可迁移点与不可迁移点）、六条证据纪律（不编造 / 时点必标 / 来源必标 / 一手与转述分开 / 事实与推断分开 / 独立信源数才是热度参考）。触发词：对标爆款、扒一下、采风、找资料、收集信息、热点、爆款分析、参考这部、题材调研、赛道调研、市场调研。本技能为套装成员，完整全流程从总控 short-drama-director 进入；〇·1 素材来源锁由总控选定。Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Script Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能 · 赛道① 国内剧】剧本创作与小说改编专精引擎：执行六阶段编剧链（来源登记→IP 与角色开发→全剧开发圣经→分集规划→正式剧本→五席圆桌审核＋授权后精修/重写），把小说、网文、大纲或已有剧本改编为来源可追溯、分集完整、台词可演的专业短剧/漫剧文字剧本。**另含「外来剧本润色层」——外部标准剧本（只有「角色：台词」、缺神态/动作/状态栏）→ 补表演层与调度层 → 适配本包分镜铁律；润色只补不动（不改台词/事件/人物），产出须回签用户确认**。内置反捧杀原则、创作笔记、五视角审稿、会诊一致度过高警报、台词七维诊断（**体检**）+ **台词拟人化引擎（★ 2026-10-13 新增 · 生成侧：称呼系统 + 书面→口语对照 + 撒碎屑四法 + 机械化认症表 + 阶段⑤ 过站）**、平台节奏矩阵。触发词：写剧本、改编小说、小说转剧本、剧本改编、分集、分集规划、开发圣经、故事圣经、IP开发、剧本审核、圆桌审...Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Storyboard Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】文戏/日常戏分镜专精引擎（短剧视感版）：短剧视感机位体系 + 镜头经济性 + 表演克制 + 段内衔接5维度与运镜5模式，将小说/剧本原文转化为极简文字分镜脚本；**并产出故事板图（★标准出图资产）：每段一张多面板图，按动作相位七态拆格（起势/过程/接触/顶点/结果/反应/出口），格数由动作复杂度决定而非机械九宫格，出口帧=下段入口帧，CLEAN_GRID 干净版供生图 + 可选技术标注版**。**★ 须感知 I 画风锁影像基底（真人/2D/3D/水墨）并按其适配脸部特写角度——3D/仿真人禁「特写×平视正面」组合、改带角度侧视（真人守 1/3 配额、2D 同场连用 ≤2）**。触发词：分镜、文字分镜、电影分镜、生成分镜、帮我分镜、导演分镜、镜头脚本、storyboard、镜头衔接、运镜衔接、故事板、故事版、分镜图、故事板图、面板链、四宫格故事板、四格故事板、六宫格故事板、九宫格故事板、2×2 故事板...Votes: 0GitHub stars: 5
- Manju Laoli Skill Sd Xuanhuan Engine【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.27 套装 · 子技能】玄幻仙侠打戏专精引擎——负责玄幻修仙/仙侠斗法/神魔异能/高燃国漫的打斗分镜与投喂提示词生成：黄金打戏颗粒度（2~3秒整秒细切·一镜一动一反馈）、双向攻防力量互换（防沙包化）、终结收束四大空间闭环协议、尺度四件套与画面几何、13种三层复合元素、法武合一招式总库、写实/修仙双轨物理参数库。触发词：玄幻打斗、仙侠斗法、法术对轰、万剑归宗、剑阵、法阵、法相、神魔大战、大招终结技、仙侠修真。套装成员：总控 short-drama-director（锁系统与投喂格式权威）+ sd-decision-engine（决策）+ sd-script-engine（剧本）+ sd-asset-engine（资产）+ sd-storyboard-engine（文戏分镜）+ sd-combat-engine（通用打戏）+ sd-xuanhuan-engine（玄幻打戏·本技能）+ sd-real-engine（真人写实）+...Votes: 0GitHub stars: 5
- Manju Laoli Skill Short Drama Director【漫剧老李AIGC 全流程Skill · V7.4.27 多技能套装版】抖音/红果爆款短剧与漫剧的工业化编剧＋视听导演系统（1 总控 + 12 子技能 · 四层纵深）。走通 剧本开发 → 分镜 → 资产提示词 → 打戏 → 投喂稿 → 出稿质检 全链路，产出可直接投喂 Seedance 2.5 / 2.0 的视频提示词组（模型版本强制锁定）。⚠️ 动手前六道前置锁（① 赛道分流 ② 素材来源 ③ 模型版本 ④ 画幅 ⑤ 交付形态 ⑥ 段缝衔接）＋ 画风锁：任何一道未指定时，必须停下询问用户并把选项摆出来让他勾，禁止默认跳过、禁止替用户推断——这是硬门，不是建议；另含动作强度三档定档。触发场景：写短剧剧本、小说改编、分集大纲、分镜脚本、九宫格场景图与人物立绘提示词、打戏分镜、方言配音标注、解说旁白稿、投喂稿生成与出稿自检。用户直接说需求即可，总控自动路由到子技能，无需了解分工（但锁必须问）。Votes: 0GitHub stars: 5
- Manjuforge漫剧/短剧连载创作引擎：题材或原著到剧本、分镜、配音文案、AI提示词全套产出，跨集记忆防人设崩坏与伏笔丢失。创作漫剧、短剧剧本、连载分镜、网文改编漫剧时使用。Votes: 0GitHub stars: 5
- Marketing VideoMakes a finished marketing video for any brand or product (feature launch, product ad, social loop) in a precise motion-design style: one morphing UI shape, a real cursor, springs, a camera that zooms, a soundtrack cut on the beat, rendered to MP4 from code. Starts by collecting the person's own context (subject, UI screenshots, design system, inspiration video), so anyone can use it. Use for '/marketing-video', 'make me a marketing video', 'a video like this one', 'fais-moi une vidéo marketi...Votes: 0GitHub stars: 5
- Micro Drama Skills Generate Media短剧媒体生成技能。根据已生成的单部作品目录，调用 Google Gemini API（单一可配置图片模型生成角色图/分镜图），将生成的图片存放到对应各集目录下。支持视觉风格预设配置。关键词：图片生成、Gemini、Google API、分镜图片、角色参考图、media generation、视觉风格。Votes: 0GitHub stars: 5
- Micro Drama Skills Produce Anime短剧制作技能。用于生成完整短剧作品，包括剧本编写、角色设计、9宫格分镜、故事板配置。每次运行生成1部作品（25集，每集30秒=上下两部分各15秒，每部分9宫格分镜）。关键词：短剧、影视、drama、剧本、分镜、storyboard、角色设计。Votes: 0GitHub stars: 5
- Micro Drama Skills Produce MvMV制作技能。根据歌曲歌词或已有MV剧本，结合用户提供的角色/场景参考图片（支持读取指定文件夹），生成完整MV分镜、故事板配置和Seedance视频生成任务。关键词：MV、音乐视频、Music Video、分镜、歌词、storyboard、视频生成。Votes: 0GitHub stars: 5
- Micro Drama Skills Submit Anime Project整部作品提交技能。读取全剧 seedance_project_tasks.json，按 Seedance 任务提交协议批量推送到 /api/tasks/push，并生成提交报告。关键词：提交任务、Seedance、批量推送、整部作品、task submit、api/tasks/push。Votes: 0GitHub stars: 5
- Minimax H3 RemoteGenerate MiniMax H3 videos from local computer by sending prompts and local Picture, Video, and Audio references to the user's authenticated MiniMax H3 Colab Remote API. Also upscale existing videos with VOSR2.Votes: 0GitHub stars: 5
- Mos Ski Motion VideoMotion Video. Make product presentations, launch reels, feature demos and event promos as motion graphics rendered from code (Remotion), from a Figma design link, a product URL or a brief. House style is simple, punchy "jitter" motion (on-twos timing, snaps, montage bursts) or calm premium. Use when the user asks for a motion video, product video, launch video, showreel, animated explainer, motion ad, or "animate this Figma".Votes: 0GitHub stars: 5
- Motion 3dMake a looping 3D motion-graphics film, 6 to 20 seconds, from a short brief. Real extruded, bevelled 3D type and SVG logos, plus boxes, spheres, tori, cylinders, cones and capsules, lit by a key light with soft shadows on a seamless floor, turning, flipping, hopping and orbiting on springs under a moving camera. Claude writes the film as one self-contained HTML file (a small WebGL2 renderer, no libraries) driven by a pure seek(t), checks it with an automated critique, and renders it to MP4, G...Votes: 0GitHub stars: 5
- Motion CaptionsMake a captioned clip, 15 to 60 seconds, from the user's own voice-over, podcast cut, interview or talking-head audio and its transcript. The words appear in time with the speech, as on Reels, TikTok, Shorts and LinkedIn, where most people watch muted - pop captions (1 to 3 big words, each popping as it is said), a karaoke line with a moving highlight, classic subtitles, or an audiogram with a speaker card and bars driven by the voice. Claude writes one HTML canvas file driven by a pure seek(...Votes: 0GitHub stars: 5
- Motion CartoonMake a looping cartoon of an original mascot or character, 6 to 20 seconds, from a short brief. A bean-shaped rig with rubber-hose arms and an antenna walks, hops, blinks, waves, reacts, talks in a bubble and holds up a sign with the product's name, in outlined flat colour with cel shading, on twos with a boiling line, squash and stretch, follow-through and a sunburst stage. Claude writes the film as one self-contained HTML canvas file driven by a pure seek(t), checks it with an automated cri...Votes: 0GitHub stars: 5
- Motion ChartsMake a looping animated data story, 6 to 20 seconds, from a brief and its numbers. A key metric counts up to its exact figure, bars and columns grow from zero with labels counting on the same spring, lines draw on, donuts sweep, a note points at the peak, the source always on screen. Claude writes the film as one self-contained HTML canvas file driven by a pure seek(t), binds every number to sourced data, runs a critique that fails dishonest charts (truncated axes, overshooting values, missin...Votes: 0GitHub stars: 5
- Motion DesignCode-only motion design pipeline (no After Effects) for Raphaël / Howseen: launch films, showreels, product promos, LinkedIn/X videos, meme clips. Use when asked to make a motion design video, a promo/launch film, a showreel, \"remake this video\", a video from a prompt (e.g. \"make a 15s motion graphics video…\"), to change a video's music/SFX, or to prepare memes for posts. Covers the brief → beat map → stills → seek(t) HTML engine → Playwright render → ffmpeg → music/SFX → QA flow, plus as...Votes: 0GitHub stars: 5
- Motion GlobeMake a data globe film, 20 to 40 seconds, of a network or a launch, worldwide. A dotted Earth from real geography forms out of the dark; the camera dives, hops and flies along great circles; routes arc between cities; pins rise as beams with HUD cards that type a query and count up their numbers; a region title, telemetry, a running total; the globe recedes into the lockup. Claude writes one self-contained HTML canvas file driven by a pure seek(t), checks every place against its country and e...Votes: 0GitHub stars: 5
- Motion Graphics Music Video NoimageCreate and revise beat-synced motion-graphics music videos from a supplied song and creative prompt, drawn entirely with the bundled native Swift renderer (vector line art, type, Metal shaders, neon glow), with music mapping, researched storyboards, procedural Swift SFX and Swift VFX. Use for full music videos or revisions to their scenes, animation, sound and effects.Votes: 0GitHub stars: 5
- Motion Launch VideosMake a kinetic-typography launch bumper or brand identity sting, a 6 to 20 second looping video of big animated type, from a short brief (product, message beats, palette, formats, duration). Claude writes the film as one self-contained HTML canvas file driven by a pure seek(t), laid out for 1:1, 9:16 and 16:9 from one config, then renders it to MP4 with real subframe motion blur, a verified seamless loop, a preview GIF and a poster. Use it for a launch video, product bumper, promo loop, anima...Votes: 0GitHub stars: 5
- Motion OverlaysMake an animated overlay for editors and streamers to lay over their own footage, delivered with alpha: a lower third (name and role, a place, a quote's attribution), a title or chapter card, social handles and a subscribe button, a YouTube end screen round the platform's elements, a corner bug, a callout pointing into the footage, a progress bar or a LIVE badge. Each comes in on springs with motion blur, holds as long as asked and goes out. Claude writes it as one self-contained HTML canvas ...Votes: 0GitHub stars: 5
- Motion ParticlesMake a particle motion-graphics loop, 8 to 15 seconds, from a short brief. Thousands of particles drift as a starfield, swarm into a word, the user's logo or a mark, burst into streaks and reform, then dissolve; spheres and grids too. Claude writes the film as one self-contained HTML canvas file driven by a pure seek(t), checks it with an automated critique (whether each particle word reads at feed size, the live area, the seam), and renders it to MP4, GIF and poster, square, vertical or wide...Votes: 0GitHub stars: 5