All authors
Claude Skills by Wanfeng1028
github.com/Wanfeng1028590 skills0 installs0 views
- Openmontage Screen Demo Publish Director屏幕演示流水线的发布导演技能，负责撰写可检索的标题与关键词、把脚本章节转成章节标记、规划缩略图概念并按平台打包发布物。适用于演示视频完成渲染后需要整理元数据与导出目录的收尾阶段。Votes: 0GitHub stars: 5
- Openmontage Screen Demo Scene Director为已有屏幕录制规划镜头与注意力路径，决定何时保持全景、何时局部放大、何时添加极简引导，适用于录屏演示的镜头编排与裁剪策略设计阶段。Votes: 0GitHub stars: 5
- Openmontage Screen Demo Script Director将已检视的录屏素材转为带时间戳的流程化脚本，让旁白与观众实际看到的操作同步，适用于录屏演示的脚本撰写与节奏标注阶段。Votes: 0GitHub stars: 5
- Openmontage Screen Recording提供屏幕录制视频的完整制作规范，涵盖录制分辨率与帧率、IDE 界面设置、光标管理、缩放平移、变速与死气剪辑、音频处理，适用于代码演示、软件教程、终端操作等录屏类内容的后期加工。Votes: 0GitHub stars: 5
- Openmontage Seedance Prompting介绍 Seedance 2.0 提示词写法，覆盖八要素结构、多镜头模式、口型同步与参数配置，适用于电影感预告、混剪与运动主导片段创作。Votes: 0GitHub stars: 5
- Openmontage Short Form定义竖屏短视频（TikTok/Reels/Shorts）的制作规范，涵盖安全区、上传规格、时长策略、1 秒钩子、节奏控制、字幕与音频标准、结构模板，适用于 9:16 短视频的策划与成片。Votes: 0GitHub stars: 5
- Openmontage Skill Creator当现有技能无法覆盖某个可复用需求时，动态创建新技能并注册到技能库。适用于管线运行中发现技术空白、新工具无使用技能或跨阶段反复出现的模式需要固化时。Votes: 0GitHub stars: 5
- Openmontage Sora Prompting讲解 Sora 2 提示词写法，包含散文加摄影块模板、进阶电影参数与调色技巧，适用于追求镜头规格精确控制的短片与广告创作。Votes: 0GitHub stars: 5
- Openmontage Sound Design面向视频制作的音频混音与声音设计规范，涵盖对白、音乐、音效的电平控制、闪避处理、平台响度目标及AI语音处理链，适用于解说视频、短视频和播客的后期混音场景。Votes: 0GitHub stars: 5
- Openmontage Stock Sourcing Usage指导如何在 OpenMontage 中高效使用 Pexels 和 Pixabay 图库工具进行素材搜索与下载，涵盖查询构建、提供商选择、许可证注意事项及资产管道集成，适用于需要免费图片和视频素材的视频制作场景。Votes: 0GitHub stars: 5
- Openmontage Storytelling为解说视频提供叙事结构模板、钩子类型、节奏控制和认知科学原则，涵盖解释弧线、误解优先法、引导发现法及Mayer多媒体学习原则，适用于动画解说和知识科普类视频的脚本创作。Votes: 0GitHub stars: 5
- Openmontage Subtitle Sync将转录数据转换为精确计时的字幕文件，覆盖计时策略、格式化与可读性，适用于竖屏和横屏视频的字幕生成与烧录。Votes: 0GitHub stars: 5
- Openmontage Talking Head Asset Director为口播视频生成字幕、音频、背景音乐与叠加图形等配套素材，并汇总为素材清单。适用于已有分镜方案与脚本、需要批量产出并校验支撑素材的口播制作阶段。Votes: 0GitHub stars: 5
- Openmontage Talking Head Compose Director按剪辑决策渲染口播成片：套用增强链、烧录字幕与叠加层、混音并编码到目标平台规格。适用于素材与剪辑方案齐备、需要产出可交付成片并完成质检的合成阶段。Votes: 0GitHub stars: 5
- Openmontage Talking Head Edit Director为口播视频编排剪辑决策表：以完整素材配合字幕叠加与可选增强为主线，规划静音处理、字幕、音频与叠加层。适用于分镜与素材清单就绪、需要确定剪辑结构再交付合成的阶段。Votes: 0GitHub stars: 5
- Openmontage Talking Head Executive Producer口播视频流水线的执行制片技能，串行调度七个阶段导演并在每个关卡做质量判断，适用于需要跨阶段状态管理、字幕同步校验与返工控制的说话人头视频项目。Votes: 0GitHub stars: 5
- Openmontage Talking Head Gen Usage指导如何使用 talking_head 工具从单张照片和音频生成说话头像视频，涵盖模型选择、参数调优、预处理模式及常见工作流，适用于虚拟形象口播、多语言本地化和快速社媒内容制作。Votes: 0GitHub stars: 5
- Openmontage Talking Head Idea Director口播视频流水线的创意阶段技能，负责探测原始素材、评估内容并产出 brief 产物，适用于已有真人讲话素材、需要确定内容定位与制作目标的起始环节。Votes: 0GitHub stars: 5
- Openmontage Talking Head Publish Director口播视频流水线的发布阶段技能，基于渲染报告生成平台元数据、缩略图概念与导出包，适用于成片完成后准备发布物料与发布记录的收尾环节。Votes: 0GitHub stars: 5
- Openmontage Talking Head Scene Director口播视频场景导演技能，分析原始素材与转录文稿，规划视觉叠加层、重构裁剪与剪辑方案，产出可渲染的场景计划。适用于需要将纯口播素材升级为视觉丰富成片的制作流程。Votes: 0GitHub stars: 5
- Openmontage Talking Head Script Director口播视频脚本导演技能，将原始口播素材转录并结构化为带时间戳分段的脚本产物。适用于已有 Brief 和素材、需要提取并整理现有语音内容的制作流程。Votes: 0GitHub stars: 5
- Openmontage Taste Direction定义剧坊视频的品味档案契约，在确定视觉识别、情绪板、提案包、自定义剧本或品牌向视频之前使用，输出可贯穿提案、场景规划、素材生成、剪辑与合成的紧凑品味画像。Votes: 0GitHub stars: 5
- Openmontage Typography面向视频制作的排版规范技能，规定字体选择、字号、安全区、字幕样式、动画时序与对比度要求，适用于在合成与素材阶段为 1080p/4K 及竖屏短视频生成文字图层时使用。Votes: 0GitHub stars: 5
- Openmontage Upscale Usage视频与图像超分辨率放大技能，说明 Real-ESRGAN 模型选择、倍率、人脸增强与降噪参数，适用于素材阶段把低分辨率或 AI 生成画面提升到目标视频分辨率时使用。Votes: 0GitHub stars: 5
- Openmontage Veo Prompting介绍 VEO 3.1 / VEO 3 提示词写法，覆盖十四要素结构、镜头运动分类与光学效果，适用于需要原生对白与精确运镜的写实或艺术风格短片。Votes: 0GitHub stars: 5
- Openmontage Video Edit面向视频剪辑日常任务的 ffmpeg 命令速查手册，覆盖裁剪、转码、缩放、变速、音视频处理、拼接、水印、画中画、滤镜、字幕、防抖、调色、压缩、幻灯片、章节标记、响度归一化等常见操作。Votes: 0GitHub stars: 5
- Openmontage Video Editing口播类视频的剪辑决策技能，规定应删与应留的内容、切点技巧、节奏控制与 edit_decisions 结构，适用于为访谈、讲解、口播素材做最终剪辑时使用。Votes: 0GitHub stars: 5
- Openmontage Video Gen Prompting视频生成提示词的通用编写技能，给出跨模型的五要素公式、镜头与运动词汇、灯光、风格、时间特效与禁忌项，适用于为各类文生视频/图生视频模型撰写提示词时使用。Votes: 0GitHub stars: 5
- Openmontage Video Reference Analyst当用户给出参考视频 URL 或本地文件并希望做出"类似这个"的作品时，本技能负责分析参考、审计能力、提出差异化方案，并先出样片再进入流水线。Votes: 0GitHub stars: 5
- Openmontage Video Stitching将多段视频素材拼接为统一成片的策略指南，涵盖顺序拼接、空间合成、AI片段链接与混合拼接四种模式，适用于叙事组装、多镜头汇编、LTX-2短片段串联以及画中画评论等场景。Votes: 0GitHub stars: 5
- Openmontage Video Understand Usage指导如何在剧坊流程中调用 video_understand 工具进行画面描述、质量评估、定向问答与场景分类，适用于渲染后质量门控、素材预分析、高光选取及生成资产校验等场景。Votes: 0GitHub stars: 5
- Openmontage Voice Performance Director让 TTS 生成的旁白听起来像被"导演"过而非"朗读"，从剧本到素材生成携带具体的语音表演计划，并通过样片进行验证。Votes: 0GitHub stars: 5
- Openmontage Whisperx基于 faster-whisper 和 WhisperX 的语音转文字工具，提供词级时间戳、语言检测和可选的说话人分离，是所有转录依赖工作流的入口。Votes: 0GitHub stars: 5
- OpenvidsMake a product-launch motion-graphics video (Claude Code launch-video style) with Remotion, a procedural soundtrack, and a fan-out of scene, sound, and design subagents. Use when the user asks for a launch video, feature announcement video, promo, explainer, or motion graphics about a product.Votes: 0GitHub stars: 5
- Opus Promo Director서비스·앱·책·제품·행사·모집의 짧은 홍보·소개 영상(10~60초, 쇼츠·릴스·유튜브)을 감독처럼 기획하고 끝까지 만든다. 사용자가 원하는 룩을 먼저 찾아 고정하고, 박자에 맞춘 Remotion 모션그래픽, AI 이미지, Three.js·Blender로 제작한 뒤 스스로 비평하고 9:16·16:9·4:5로 납품한다. 기존 영상에 자막만 넣거나 자르고 붙이는 편집에는 쓰지 않는다.Votes: 0GitHub stars: 5
- Opus Visual Motion EngineBuild motion graphics, showreels, brand/product/explainer videos, kinetic type, landing pages, dashboards, mockups, diagrams and data viz as single-file HTML: real brand material, a named direction, deterministic seek(t) timing, designed cut-to-cut handoffs, a composed score, render QA and MP4 export. Use to make, animate, render or review a video, animation, UI or visual. Triggers: video, vídeo, animação, showreel, motion graphic, explainer, landing page, dashboard, mockup, diagram.Votes: 0GitHub stars: 5
- Painted AnimationMake hand-painted watercolour-and-ink cartoon videos (MP4) with code — p5.js + p5.brush rendered frame by frame in headless Chrome, encoded with ffmpeg — starring Clawd or any character. Use when the user asks to make an animation, animated short, cartoon, music video, lyric video, animated GIF/loop or "a video of Clawd…", or mentions the PDoomVideo / ClaudeAnimationBase approach. Covers storyboarding, the engine API, music and LRC lyric sync, cloud-friendly preview review, and choosing local...Votes: 0GitHub stars: 5
- Parsing VideoInterpret video content visually by sampling frames into timestamped contact sheets that can be read as images. Use when: user asks what happens in a video; asks to summarize, describe, review, or QA video content or footage; asks about scenes, actions, people, or objects in a video; needs a storyboard-style overview of a clip; asks to find where something occurs in a video. Triggers on 'watch this video', 'what's in this video', 'summarize the video', 'describe the footage', 'contact sheet',...Votes: 0GitHub stars: 5
- Personal Brand CreatorPersonal brand and creator video: vlogs, tutorials, opinion pieces, day-in-the-life, and channel content. Authentic pacing, jump cuts, talking-head craft, and a mandatory asset-request protocol. Use for creators, coaches, founders, and solo brands.Votes: 0GitHub stars: 5
- Pharma Medical DevicePharma and medical-device video: mechanism-of-action explainers, trial-data visualisation, device instructions-for-use, HCP education and patient-support content. Regulation-aware, balanced, accurate, with strict approved-wording, fair-balance and adverse-event handling and a mandatory asset-request protocol. Use for pharmaceutical, biotech and medtech companies.Votes: 0GitHub stars: 5
- Podcast Talking HeadPodcast and talking-head editing with a mandatory visual-narration layer: multi-cam switching, spoken-word pacing, motion-graphic cutaways and B-roll synced to the transcript, lower-thirds, and level discipline. Requires the client's actual recording. Use for podcast episodes, guest clips, interviews, and talking-head content.Votes: 0GitHub stars: 5
- Premiere Ffmpeg Render영상 편집을 프리미어 없이 FFmpeg로 하는 스킬. (A) 프리미어 프로(.prproj) 프로젝트를 분석해 컷 목록·효과(위치/크기, 자르기, 울트라 키 크로마키, 마스크 모자이크, 화면 정지, 트랙 음소거)를 뽑고 같은 영상을 다시 렌더링한 뒤 원본 내보내기와 화질(PSNR)·소리 싱크를 비교 검증한다. (B) 프로젝트 없이 원본 녹화(화면녹화+그린스크린 카메라)만으로 처음부터 편집한다 — 음성 인식으로 대본을 만들어 다시 말한 부분·끊긴 문장·군소리·긴 침묵을 자르고, 카메라를 키잉해 얹고, 화면 속 개인정보를 모자이크하고, 결과를 다시 받아써서 검증한다. 사용자가 "프리미어 프로젝트 분석해줘", "prproj 읽어줘", "이 편집 그대로 다시 만들어줘", "프리미어 없이 렌더링", "편집 재현", "컷 목록 뽑아줘", "원본만 가지고 편집해줘", "이 녹화본 편집해줘", "말 반복한 거 잘라줘", "NG 잘라줘", "컷 편집", "설명 영상 만들어줘", "그린스크린 합성...Votes: 0GitHub stars: 5
- Preset 001 Zayyan Blue GlassPreset 001 'Blue Glass' - a full editing skill locked to one captured style: the measured royal-blue glassmorphism portfolio-reel look (soft blue gradients, floating glass UI cards, dotted connectors, the speaker composited into the scene, kinetic mapped text). Same structure and depth as every skill in this pack, with the look, motion and pacing pre-decided from a measured reference. Use when a client wants this exact soft-blue premium reel.Votes: 0GitHub stars: 5
- Preset 002 Realtor Word CaptionPreset 002 'Realtor Word-Caption' - a full editing skill locked to one captured style: a bright, high-key vertical talking-head reel with word-by-word bold captions, a two-colour keyword accent system (blue/cyan + orange/amber), hard cuts plus a whip/motion-blur transition, and a frosted-glass pill end card. Same structure and depth as every skill in this pack, with the look, motion and pacing measured from a reference. Use when a client wants this exact realtor / creator talking-head reel look.Votes: 0GitHub stars: 5
- Preset 003 Saaswave Tactile PurplePreset 003 'SaaSWave Tactile Purple' - a full editing skill locked to one captured style: a bright, high-key 3D 'digital workspace' product reel - photoreal 3D office objects in a purple/magenta wash, floating translucent UI cards, a purple triangle cursor, 3D buttons, and a fluid morph from type into a cursive logo. Measured from a reference. Use when a client wants this exact tactile-purple SaaS/motion-design product look.Votes: 0GitHub stars: 5
- Preset 004 Cloudy Esport NeonPreset 004 'Cloudy eSport Neon' - a full editing skill locked to one captured style: a high-energy vertical eSports montage - futuristic blue-neon architecture, silhouettes, heavy glow/bloom, glitch and whip transitions, layered skewed type, 3D milestone numerals, a chat-bubble UI and a gold trophy. Measured from a reference. Use when a client wants this exact eSports hype look.Votes: 0GitHub stars: 5
- Preset 005 Higgsfield Dark UiPreset 005 'Higgsfield Dark-UI' - a full editing skill locked to one captured style: a high-contrast product/software demo - dark-mode UI floating in a black void, 2D-to-3D spatial UI transforms, a neon lime-green accent, a neon-blue command line, cursor micro-interactions, glitch typography, and a light-mode end card. Measured from a reference. Use when a client wants this exact dark-UI AI/software launch look.Votes: 0GitHub stars: 5
- Preset 006 Podcast Overlay ExplainerPreset 006 'Podcast-Overlay Explainer' - a full editing skill locked to one captured style: a single static talking-head in a dark, desaturated studio carrying a bright graphics layer - UI-mimicking overlays, full-screen cutaway diagrams, one yellow keyword, white drop-shadow captions, a brand end-card. Measured from a reference. Use when a client wants this exact 'podcast framing + explainer engine' format.Votes: 0GitHub stars: 5
- Preset 007 Creator Kinetic TextPreset 007 'Creator Kinetic-Text' - a full editing skill locked to one captured style: a stark white studio, a single presenter in a black tee, kinetic word-by-word text overlaid on the chest, an orange script for connective phrases, orange pill badges, outlined grey section numerals, screenshot/UI proof cutaways, a colour-inverted emphasis frame, and a brand end-card. Measured from a reference. Use when a client wants this exact high-energy creator / 'infotainment' look.Votes: 0GitHub stars: 5
- Preset AuthoringTurns a reference video (supplied as a frame-level contact-sheet PDF, a video file, or a link) into a reusable PRESET: a named, portable style recipe covering palette, type behaviour, motion grammar, transitions, pacing and structure, saved as a skill file under presets/. Use when the user supplies reference frames or a video and wants that exact editing style captured and reproduced.Votes: 0GitHub stars: 5