{"skills":[{"name":"asmr-ambient","version":"0.1.14","hash":"sha256:4c8fcba3766b4c7f42d9f95673746b55e1750174bdd7b3228af9ca9c595497fc","summary":"Generate sleep meditation and ASMR audio","summaryZh":"生成助眠冥想与ASMR音频","description":"Sleep & Relaxation Audio Creator. Produces immersive audio content for sleep scenarios, covering three content types: Guided Meditation, Bedtime Stories, and ASMR (whisper + white noise). Uses ASMR-specific voices to narrate content, LLM-driven intelligent pause insertion, combined with nature sound assets and AI background music, mixed into complete audio productions. Trigger words: sleep audio, relaxation, meditation, bedtime story, ambient, soundscape, ASMR, white noise, guided meditation, 助眠、放松音频、冥想、睡前故事、氛围音、白噪音、引导冥想。","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"助眠音频","tagEn":"Audio Content / Audio Generation","tagCn":"音频内容 / 音频生成","completeTagsEn":["Sound & Music / Creative Generation"],"completeTagsCn":["音频音乐 / 创作生成"],"descEn":"For sleep, meditation and relaxation audio production covering guided meditation, bedtime stories and ASMR content types. Takes a scene, theme or content direction as input. Produces fully mixed audio combining ASMR-specific TTS with whisper characteristics, intelligent pacing with natural breath pauses, curated nature sound assets and AI-generated background music. Optimized for immersive listening experiences.","descCn":"适用于助眠、放松、冥想类音频内容的制作,覆盖引导冥想、睡前故事、ASMR(耳语加白噪音)三种内容类型。输入场景或内容主题,输出混音完成的成品音频。制作流程组合 ASMR 专用耳语音色、带自然呼吸间隔的智能停顿节奏、精选自然音效素材与 AI 生成的背景音乐,针对沉浸式聆听体验进行优化。","tools":[],"skillType":"skill","sortWeight":10,"downloads":2509,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"audiobook","version":"0.3.16","hash":"sha256:56a7234033d1dcb452628881d4c3be1c5073177e98a0ed586c966babf5c16c79","summary":"Turn books into multi-voice audiobooks","summaryZh":"把书籍转化为多角色有声书","description":"Audiobook creation assistant. Converts book text into multi-character narrated audio, supporting audiobook production, multi-character voiceover, novel narration, TTS voiceover, and read-aloud scenarios. Automatically identifies dialogue and narration, assigns a distinct voice to each character, intelligently adds pause markers, and generates natural, fluent audiobook audio. Trigger phrases: audiobook, read aloud, TTS book, multi-character voiceover, novel narration, book narration, voice acting, narration, 有声书, 朗读, 读书, 多角色配音, 小说朗读, 读书配音. Supports chapter-level generation — user confirms the first chapter, then remaining chapters continue sequentially.","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":["audiobook","narration","read aloud","text to audiobook","TTS book","multi-character voiceover","novel narration","book narration","voice acting","有声书","朗读","读书","多角色配音","小说朗读"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"有声书","tagEn":"Audio Content / Audio Generation","tagCn":"音频内容 / 音频生成","completeTagsEn":["Sound & Music / Creative Generation"],"completeTagsCn":["音频音乐 / 创作生成"],"descEn":"For converting books into multi-character narrated audio with professional-grade production quality. Takes book text typically chapter by chapter as input material. Produces audiobook audio with automatic dialogue and narration detection, per-character voice assignment for distinct character voices, smart pause insertion at natural break points and chapter-by-chapter generation. Supports both fiction and non-fiction genres.","descCn":"适用于将文本书籍转换为多角色配音有声音频的场景,产出专业级制作品质的有声书。输入书籍文本(通常分章节处理),输出有声书音频,自动识别对话与旁白段落、为每个角色分配独立且有辨识度的音色、在自然停顿点智能添加间隔、按章节顺序逐步生成。同时支持虚构类与非虚构类书籍体裁。","tools":[],"skillType":"skill","badges":["trending"],"sortWeight":20,"downloads":16166,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/c9d8e5e2-1370-40b4-87da-dde20f536cb9.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"clip-export","version":"3.0.13","hash":"sha256:5d9e6663e813ff426c02f2ccc3b8fd6e0c62d58a44200f1ed795d839d7bc27ab","summary":"Push videos to JianYing or CapCut drafts","summaryZh":"把视频推送到剪映或CapCut草稿","description":"Video project push tool. Programmatically creates editing projects via Python scripts, with one-click push to JianyingPro or CapCut. Direct mode creates project files directly in the local JianyingPro/CapCut draft directory — open the app and start editing. Built on pyJianYingDraft (PyPI). Trigger words include: JianyingPro, CapCut, push, push to JianyingPro, capcut, export to JianyingPro.","tags":["Platform Tooling"],"tagsCn":["平台工具"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"剪映导出","tagEn":"Visual Production / Post-Production","tagCn":"视觉创作 / 后期制作","completeTagsEn":["Platform Tooling"],"completeTagsCn":["平台工具"],"descEn":"For pushing AI-assembled video projects into JianYing or CapCut as editable drafts for manual post-production. Takes a video project including clips, timeline and assets as input. Produces a draft directly in the local JianYing or CapCut draft directory ready to open and edit in the desktop application. Built on the pyJianYingDraft library for reliable draft format compatibility and seamless workflow integration.","descCn":"适用于把已组装好的视频工程推送到剪映或 CapCut 进行人工后期精修的场景。输入视频工程(包含片段、时间轴与素材资源),输出直接生成在本机剪映或 CapCut 草稿目录的工程文件,打开桌面应用即可编辑。基于 pyJianYingDraft 库封装,确保草稿格式兼容性与工作流的无缝衔接。","tools":[],"skillType":"skill","sortWeight":30,"downloads":13735,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"drama-soundtrack","version":"0.1.15","hash":"sha256:2ef94e07dc28538c996c2237fe08dce23ef97ba296b9f75e198b7bc35af2b1a6","summary":"Compose character themes and scene BGM","summaryZh":"创作角色主题曲与场景配乐","description":"Drama original soundtrack creation assistant. Creates complete soundtrack suites for TV series, films, and short dramas, including character theme songs, opening themes (OP), ending themes (ED), and scene background music (BGM). Deeply analyzes the plot based on scripts and character biographies, creating custom theme songs for each main character, composing opening and ending themes for the entire show, and crafting background music for key dramatic moments. Supports importing scripts from the short-drama-screenwriter skill, as well as external scripts and character biographies. Trigger words include: soundtrack, score, theme song, opening theme, ending theme, BGM, OST, drama soundtrack, film score, series score, character theme, character theme song, compose music for a drama, film and TV music, create a theme song, create an ending theme, climax BGM.","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"影视配乐","tagEn":"Audio Content / Audio Generation","tagCn":"音频内容 / 音频生成","completeTagsEn":["Sound & Music / Creative Generation"],"completeTagsCn":["音频音乐 / 创作生成"],"descEn":"For composing a complete original soundtrack package for film, TV and short drama productions. Takes a script and character bios as input, optionally imported from short-drama-screenwriter skill or supplied externally. Produces character theme songs, opening and ending themes and key-scene BGM tracks, all derived from deep plot analysis. Each composition reflects character arcs and narrative tension for authentic dramatic scoring.","descCn":"适用于为电视剧、电影、短剧创作完整原声音乐套件的场景,提供专业级影视配乐方案。输入剧本与人物小传(可从 short-drama-screenwriter 技能导入或外部直接提供),输出角色主题曲、片头曲、片尾曲与关键场景 BGM 配乐。所有音乐基于剧情深度分析创作,每首作品反映角色弧线与叙事张力,实现真实感的戏剧配乐效果。","tools":[],"skillType":"skill","sortWeight":40,"downloads":2580,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"dynamic-poster","version":"0.5.14","hash":"sha256:b5db0290dbfe66f4eda5f6bb6cf815fe9fe1098ba4fb32ef3e275628306f06c3","summary":"Surreal dynamic posters from product art","summaryZh":"用产品素材生成超现实动态海报","description":"Dynamic poster / motion graphic creator for product marketing. Transforms brand assets (logo, product photos) into surreal short video posters using \"impossible juxtaposition\" — products in physics-defying scenes, shot with realism. Trigger: \"dynamic poster,\" \"motion poster,\" \"animated poster,\" \"video poster,\" \"product poster video,\" \"social media poster,\" \"event teaser,\" \"motion graphics,\" \"animated visual,\" \"brand motion design,\" \"poster animation.\" NOT for: long-form promo (use promo-video), MV (use mv-creator), static posters.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"动态海报","tagEn":"Advertising / Video Generation","tagCn":"广告营销 / 视频生成","completeTagsEn":["Commercial Ad / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成"],"descEn":"For social-media motion posters and event teasers requiring surreal visual impact. Takes brand assets such as logo, product photos and color palette as input. Produces short surreal-style video posters featuring impossible juxtapositions shot with photographic realism for social media campaigns and promotional use. Supports vertical and square aspect ratios optimized for Douyin, Instagram Stories and WeChat Moments.","descCn":"适用于社媒动态海报、活动预告、品牌运营素材的制作,支持竖版和方形两种画幅比例。输入 logo、产品图、品牌色系等品牌资产,输出超现实风格的短视频海报,以“不可能的并置”概念配合写实摄影质感呈现产品。适配抖音、Instagram Stories、微信朋友圈等主流社交平台规格,自动匹配最优画面构图与运动节奏。","tools":[],"skillType":"skill","sortWeight":50,"downloads":2135,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"image-remix","version":"0.7.16","hash":"sha256:e9060cd76237b0babad01b40ca0f925771943f57c1dc458db6a5344013783d81","summary":"Generate new images with a reference vibe","summaryZh":"复刻参考图氛围生成新图","description":"Analyze reference form, aesthetic, and mood; confirm what to preserve and replace; create variants through text reconstruction or authorized guidance. Not for editing, copying, upscaling, or video.","tags":["E-Commerce"],"tagsCn":["电商"],"creator":"","triggerWords":["remix this image","same vibe different content","make something like this","recreate this feel","style-inspired image","image remix"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"图片重混","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["E-Commerce / Creative Generation"],"completeTagsCn":["电商 / 创作生成"],"descEn":"For creators who want new images inspired by a reference without copying its subject. Users provide one or more references, describe the replacement content, choose which visual qualities to preserve, and confirm whether direct reference guidance is authorized. The workflow analyzes ten aesthetic dimensions, condenses them into Form, Aesthetic, and Mood, confirms a remix direction, selects a capability-compatible model, generates independent variants, and iterates on chosen dimensions. It delivers selected images, a reusable prompt, and a concise style summary. It is not for editing the original, exact reproduction, upscaling, format conversion, certain source-model identification, or image-to-video work.","descCn":"面向希望借鉴参考图感觉但重新创作内容的用户。用户提供一张或多张参考图、希望替换的新主体，并确认要保留的构图、色彩、光影、质感、媒介或氛围，以及是否有权直接使用参考图。流程先从十个美学维度拆解画面，再归纳为形、韵、意三个轴，确认重混方向，按所需能力选择模型，生成独立变体，并围绕指定维度迭代。最终交付选定图片、可复用提示词和精简风格摘要；不用于修改原图、精确复制、放大转换、确定识别源模型或图片转视频。","tools":[],"skillType":"skill","badges":["trending"],"sortWeight":60,"downloads":16638,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/5e198825-e672-48da-9c8f-b3cdd3165cac.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"minecraft-pixel-art","version":"0.1.13","hash":"sha256:bf2fed0851275bd277ae3cf6612ae071c90a31b5bdf778209844460b29c4ed6e","summary":"Convert images to Minecraft pixel art","summaryZh":"把图片转换为Minecraft像素方块画","description":"Turn one source image into Minecraft-inspired 16x16 voxel-block pixel art while preserving the main silhouette and identifying colors. Use for avatars, objects, scenes, and game-style social images.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"像素方块画","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"For converting any image into Minecraft-style pixel art with authentic block aesthetics and voxel representation. Takes a source image as input material. Produces a 16x16 voxel and block-style stylized version aligned with Minecraft visual language. Suitable for game-related content creation, social media posts, avatar generation and creative pixel art projects requiring the distinctive Minecraft blocky aesthetic.","descCn":"适用于把任意图片转换为 Minecraft 风格像素方块画的场景,忠实还原方块美学与体素表现。输入一张原始图片,输出 16x16 体素方块风格的像素方块画,贴合 Minecraft 美术语言。适用于游戏相关内容创作、社交媒体配图、头像生成以及需要 Minecraft 标志性方块美学的创意像素艺术项目。","tools":[],"skillType":"skill","sortWeight":70,"downloads":2305,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"mv-creator","version":"0.1.16","hash":"sha256:e24cd1c6a9ae4d548707e1876fdc7f0e7138b2909513cae66d31e91be483c557","summary":"Turn lyrics into MV storyboard and video","summaryZh":"把歌词转化为MV分镜与成片","description":"Music Video (MV) creation assistant. Triggered when a user needs to create an MV, produce music video storyboards, generate AI video prompts, plan MV timelines, or needs editing software workflow guidance.","tags":["Animation","Creative & Experimental"],"tagsCn":["动画","创意实验"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"MV创作","tagEn":"Music Video / Video Generation","tagCn":"音乐视频 / 视频生成","completeTagsEn":["Animation / Planning","Creative & Experimental / Planning"],"completeTagsCn":["动画 / 计划制定","创意实验 / 计划制定"],"descEn":"For music-video creation from lyrics or song to finished cut covering the full production pipeline. Takes lyrics or a music track plus desired style direction as input material. Produces an MV storyboard with shot breakdown, AI generation prompts for each scene, an MV timeline with beat synchronization and editing-software workflow guidance. Delivers a complete generated music video ready for review and publishing.","descCn":"适用于音乐视频从歌词到成片的全流程创作,覆盖创意构思、分镜设计到成片输出的完整制作管线。输入歌词或音乐文件、期望的视觉风格方向,输出 MV 分镜故事板(含逐镜头拆解)、各场景 AI 视频生成提示词、与节拍同步的 MV 时间线规划、剪辑软件工作流指引,最终交付可供审阅和发布的 MV 成片。","tools":[],"skillType":"skill","sortWeight":80,"downloads":3391,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/deb44193-a6e6-46c7-8d19-f602ad240201.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"skill-creator","version":"2.3.4","hash":"sha256:054ad7a1d709860098035e9f9675a3596c40c2cd3b8d86df4ac727b9aa71a715","summary":"Create bilingual reusable skills","summaryZh":"创建双语可复用Skill","description":"Create, update, install, manage, or reuse Skills from an intent, workflow, or completed conversation. Produce runtime files and metadata. Also use to repeat a completed process with new inputs.","tags":["Platform Tooling"],"tagsCn":["平台工具"],"creator":"","triggerWords":["create skill","modify skill","save as skill","save this","turn into a skill","solidify workflow","remember this approach","want to do this again","install skill","download skill","skill marketplace","manage skill","make this reusable","update skill","setup skill"],"guidePrompt":"帮我使用它来创建一个新的技能。首先询问我这个技能应该做什么。","guidePromptEn":"Help me use it to create a new skill. First ask me what this skill should do.","displayNameZh":"技能创建","tagEn":"Platform Tooling","tagCn":"平台工具","completeTagsEn":["Platform Tooling"],"completeTagsCn":["平台工具"],"descEn":"For users who need to create, revise, install, or reuse a Skill. Accepts a new task intent, an existing workflow, current Skill files, or a completed conversation. It extracts the scope, inputs, outputs, and key decisions; writes semantically aligned English and Chinese runtime files plus market metadata; adds references or scripts when useful; then checks structure, trigger wording, bilingual consistency, references, and versioning. Delivers a complete Skill directory ready to save, install, or submit to the market, together with validation results. Best for preserving repeatable multi-stage workflows; not intended for read-only quality reviews or for embedding platform tool calls and request parameters in a Skill.","descCn":"面向需要创建、修改、安装或复用 Skill 的用户。可输入新的任务意图、已有工作流、现有 Skill 文件或已完成的对话记录；流程会先提炼适用场景、输入输出和关键决策，再生成语义对齐的中英文运行时文件与市场元数据，按需补充 references 或 scripts，随后检查目录、触发表达、双语一致性、引用完整性和版本信息。最终交付可保存、安装或提交市场的完整 Skill 目录及校验结果。适合固化可重复的多阶段流程，不适合只做独立质量审查或把平台工具调用细节写入 Skill。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":90,"downloads":14835,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"2.3.4"},{"name":"skill-reviewer","version":"0.4.1","hash":"sha256:a739abd5298e5da877e310346151869709a6352446d07bb1d58f0c286cd1e631","summary":"Audit skill quality and structure","summaryZh":"审查Skill的质量与结构","description":"Skill quality review tool. Reviews SKILL.md structure, description quality, trigger-word coverage, plus whether meta.yaml display fields meet the latest Hub publishing spec (length, category enum, zh/en alignment, banned naming wording). Read-only analysis — never modifies files. Trigger phrases: review skill, audit skill, skill quality check, check skill quality, meta validation, pre-publish check, field spec check.","tags":["Platform Tooling"],"tagsCn":["平台工具"],"creator":"","triggerWords":[],"guidePrompt":"Help me review this Skill's quality: check its structure, trigger-word coverage, and best-practice compliance, and provide a score with prioritized improvement suggestions.","guidePromptEn":"Help me review the quality of this Skill: check its structure, trigger word coverage, and best practice compliance, and give a score along with prioritized improvement suggestions.","displayNameZh":"技能审查","tagEn":"Platform Tooling","tagCn":"平台工具","completeTagsEn":["Platform Tooling"],"completeTagsCn":["平台工具"],"descEn":"For auditing the quality and structure of an existing skill before publishing to the skill market. Takes a SKILL.md file or skill directory as input material. Produces a read-only quality report covering structure compliance, description quality assessment, trigger-word coverage analysis and best-practice conformance. Delivers an overall score with prioritized improvement suggestions for reaching publication standards.","descCn":"适用于发布前对已写好的 Skill 做质量审查的场景,确保达到上架标准。输入 SKILL.md 文件或 skill 目录,输出只读的质量审查报告。报告覆盖结构合规性检查、描述质量评估、触发词覆盖度分析、最佳实践合规性校验,并给出综合评分与优先级排序的改进建议,帮助作者高效提升 Skill 质量。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":100,"downloads":15635,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"0.4.1"},{"name":"coordinate-camera-control-designer","version":"0.1.6","hash":"sha256:bdc2f4574cb6bedbba3e2d76a67ac56f263afb617e8e271b9585eda2b2cbc3ec","summary":"Convert action scenes into coordinate-based camera-control prompts and storyboard planning assets.","summaryZh":"坐标化运镜与分镜规划","description":"Coordinate camera control & visual storyboarding assistant. Converts video, image-to-video, action storyboards, parkour chases, fight impacts, camera follows, and push/pull/pan/tilt requests into executable `[L,T,W,H]` coordinate prompts, and plans scene motion-path overview images, 3x3 master storyboards, and video-generation prompts. Trigger words: coordinate camera, coordinate shot control, coordinate prompt, L T W H, bbox camera, Bounding Box prompt, shot-path coordinates, character from A to B, impact-point coordinates, coordinate video control, image-to-video camera control, high-speed fight prompt, 3x3 storyboard, motion-path map. Boundary: this Skill handles coordinate direction and planning-map / storyboard generation — it does not replace long-form scripts, subtitle production, or plain single-shot generation.","tags":["Commercial Ad","Animation"],"tagsCn":["商业广告","动画"],"creator":"","triggerWords":["coordinate camera","coordinate shot control","coordinate prompt","bbox camera","Bounding Box prompt","shot-path coordinates","impact-point coordinates","image-to-video camera control","high-speed fight prompt","3x3 storyboard","motion-path map"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"坐标运镜分镜设计师","tagEn":"Commercial Ad / Planning","tagCn":"商业广告 / 计划制定","completeTagsEn":["Commercial Ad / Planning","Animation / Planning"],"completeTagsCn":["商业广告 / 计划制定","动画 / 计划制定"],"descEn":"Coordinate-based camera control and visual storyboarding. Converts subject position, movement targets, impact points, and camera follow, push, pull, pan, and tilt directives into controllable [L,T,W,H] coordinate prompts. Generates a 4K scene motion-path overview image and a black-and-white line-art 3x3 master storyboard, then outputs Seedance video camera-motion prompts with embedded coordinate control for reliable action, chase, and fight-beat execution.","descCn":"坐标运镜与可视化分镜。把主体位置、运动目标、撞击点、镜头跟随/推拉摇移转写成可控的 [L,T,W,H] 坐标提示词；生成 4K 场景运动轨迹图与黑白线稿 3x3 总分镜，输出带坐标控制的 Seedance 视频运镜提示词；适合动作、追逐、打斗、撞击等高速动作场面。","tools":[],"skillType":"skill","sortWeight":110,"downloads":2826,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/500c4998-65ac-49a5-8518-dbbb18ce1138.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"mg-voiceover-animation-generator","version":"0.1.6","hash":"sha256:d5011d570209ca278cc9239f7fed9ae72e6a6d145a7ed507c48f5930d3aefb00","summary":"Create reviewable H3 prompts for explainers, launches, demos, and feature walkthroughs before final animation generation.","summaryZh":"基于口播、产品素材或主题需求，生成 H3 提示词并产出 MG 动画，适用于讲解与演示场景。","description":"MG voiceover animation generator for product, interface, feature, launch, and abstract-theme explainers. It first creates a reviewable H3 prompt document from the user's inputs and reference materials, then produces the complete MG animation only after user approval. Supports product images, App/SaaS screenshots, voiceover scripts, audio, product materials, real products, fictional concepts, and no-reference-image scenarios. The visual style is inferred from the input rather than selected from a fixed template. Trigger phrases include: MG voiceover animation generator, MG animation, voiceover to MG, product explainer animation, feature demo animation, app demo, SaaS demo, concept product animation, abstract theme animation, H3 MG, H3 prompt, H3 animation prompt.","tags":["Education"],"tagsCn":["教育"],"creator":"","triggerWords":["MG voiceover animation generator","H3 MG animation","H3 voiceover animation","H3 product MG animation","product MG animation","MG animation","MG explainer","MG prompt","H3 prompt","H3 animation prompt","voiceover to MG","voiceover to animation","product explainer animation","product explanation animation","feature demo animation","interface demo animation","app demo animation","SaaS demo animation","product launch film","product launch animation","product introduction animation","concept product animation","abstract theme animation","H3 motion graphics","motion graphics explainer","product demo","app demo","feature demo","product launch video"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"MG 口播动画生成器","tagEn":"Education / Creative Generation","tagCn":"教育 / 创作生成","completeTagsEn":["Education / Creative Generation"],"completeTagsCn":["教育 / 创作生成"],"descEn":"Create a complete H3 MG animation prompt for explainers, launches, demos, walkthroughs, concepts, and abstract themes. The user may provide product images, screenshots, voiceover text, audio, product documents, a real product, a fictional product name, or only a theme. The skill first clarifies missing voiceover, on-screen text, duration, ratio, style, and whether a product is real or fictional when needed. It then uses MG animation references to write a H3 prompt with delivery specs, product setup, visual style, narrative structure, timed shots, screen text, voiceover, element motion, transition bridges, brand or product rules, music, and subtle sound effects. The complete prompt is placed on the canvas as a Markdown document for review, and final H3 generation starts only after explicit approval.","descCn":"面向需要制作 MG 动画讲解、产品发布、功能演示、App/SaaS 界面说明、概念产品和抽象主题表达的用户。用户可提供口播文案、配音音频、产品图、界面截图、产品资料或主题设定。Skill 会先判断素材类型、真实或虚构属性和缺失信息，再结合 MG 动画知识库确定视觉风格与叙事结构，生成包含规格、分镜节奏、屏幕文字、Voiceover、动效、转场、品牌规则和音频设计的 H3 提示词。提示词会先上画布供用户审阅，确认后才生成最终 MG 动画。适合信息讲解和产品演示，不适合真人写实视频或复杂剧情短片制作。","tools":[],"skillType":"skill","sortWeight":120,"downloads":2352,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/8bceeece-4bc1-4467-92bd-7737073ff253.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"short-drama-series-writer","version":"0.1.5","hash":"sha256:3217b7044582d1617bfe030f8379759dbb82999da410df54ce4869d827f6494a","summary":"Create full short-drama scripts from concept to episode export.","summaryZh":"一键短剧全流程剧本","description":"Full-process short-drama scriptwriting assistant. Guides the user from topic selection, target audience, character relations, pacing hooks, paywall beats, episode directory, single-episode drafting, to complete export, generating a 50-100 episode short-drama package. Fits domestic vertical-screen dramas and overseas ReelShort / DramaBox, covering god-of-war, baby-genius, comeback, sweet-love, revenge, and similar genres. Trigger words: short drama, mini drama, vertical drama, short drama script, episode outline, hook beat, paywall, ReelShort, DramaBox.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":["short drama","mini drama","vertical drama","short drama script","episode outline","hook beat","paywall","ReelShort","DramaBox"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"微短剧剧集编剧","tagEn":"Short Drama / Intent Brainstorming","tagCn":"短剧漫剧 / 意图脑爆","completeTagsEn":["Short Drama / Intent Brainstorming"],"completeTagsCn":["短剧漫剧 / 意图脑爆"],"descEn":"Full-process short-drama scriptwriting. Covers topic selection, target audience, character relations, pacing hooks, paywall beats, episode outlines, single-episode drafts, and complete export, generating a 50 to 100 episode short-drama package. Supports vertical-screen domestic dramas and overseas ReelShort or DramaBox, spanning god-of-war, baby-genius, comeback, sweet-love, and revenge genres. Produces text scripts only, does not generate video.","descCn":"微短剧全流程剧本创作。从选题、目标受众、人物关系、爽点节奏、付费卡点、分集大纲到单集剧本与完整导出，系统生成 50-100 集短剧方案与剧本；支持竖屏与出海 ReelShort/DramaBox，涵盖战神/萌宝/逆袭/甜宠/复仇等类型。产出为纯文字剧本，不生成视频。","tools":[],"skillType":"skill","sortWeight":130,"downloads":3020,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/a79c12c9-d43e-4219-a8ce-a43ececcc340.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"chinese-style-short-drama-generator","version":"1.2.5","hash":"sha256:ecc4740c34a1ebb59d3d2de88c6b5a3da3a21ce16dae69a0e24983ed0388ded3","summary":"Create full Chinese-style short drama videos from ideas or scripts.","summaryZh":"国风短剧全流程成片","description":"End-to-end Chinese-style short-drama production workflow. Applies when the user says \"turn a script into a video\", \"short-drama script to video\", \"write the script first, then storyboard and video\", \"Chinese-style short drama end-to-end\", \"short drama from script to storyboard to video\", etc. The entry point asks step by step about script status, aspect ratio, character references, and visual direction; when there is no script, the single-episode writing flow starts first; when there is a character reference, multi-angle character reference sheets are generated. Built-in Chinese-style short-drama aesthetics and full AI expression-prompt logic.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":["Chinese-style short drama","Chinese-style short-drama film","short-drama script to video","script to video","single-episode short-drama video","short-drama storyboard video","script to video","drama script to video","script to short drama video"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"国风短剧成片生成器","tagEn":"Short Drama / Creative Generation","tagCn":"短剧漫剧 / 创作生成","completeTagsEn":["Short Drama / Creative Generation"],"completeTagsCn":["短剧漫剧 / 创作生成"],"descEn":"Full workflow for Chinese-style short-drama production. Step by step confirms script status, aspect ratio, character references, and visual direction; drives from script to storyboard brief, multi-angle character sheets, scene reference sheets, 2x4 monochrome panels, panel-level video prompts, video generation, and final assembly. Includes built-in Chinese-style and xianxia short-drama aesthetics plus full AI expression and natural-blink performance logic.","descCn":"全流程国风短剧制作工作流。分步确认脚本状态、画面比例、角色参考、画风方向；脚本→故事板概要→角色多角度设定图→场景设定图→2x4 黑白分镜→Panel 级视频提示词→视频→合成成片；内置国风/仙侠短剧画风与完整 AI 表情、自然眨眼表演逻辑。","tools":[],"skillType":"skill","sortWeight":140,"downloads":2375,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/d43b013f-5a76-45f4-8f8b-c43b2d8097c0.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"brand-font-effect-editor","version":"0.1.5","hash":"sha256:2c7ab17cbc76a96521e461a7a14441d007aa7ef27e05a3fa4e7e6f5114a6897c","summary":"Transfer brand font effects from one image to another.","summaryZh":"识别并迁移品牌字效","description":"Brand font-effect transfer Skill. Identifies logo lettering, bubble fonts, handwriting, decorative type, and brand custom fonts in a reference image, then transfers either the glyph shape or the full font effect onto specified text in a target image. Produces a font-identification report first, then confirms font source, licensing version, replacement target, text content, copy scope, output size, and font size through a single selection card. Trigger words include: brand font transfer, font recreation, font replacement, logo font transfer, bubble font replacement, copy font style, replace font effect.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["brand font transfer","font recreation","font replacement","logo font transfer","bubble font replacement","copy font style","replace font effect"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"品牌字效迁移编辑器","tagEn":"Commercial Ad / Post-production","tagCn":"商业广告 / 后期处理","completeTagsEn":["Commercial Ad / Post-production"],"completeTagsCn":["商业广告 / 后期处理"],"descEn":"Brand font-effect transfer. Identifies logo lettering, bubble fonts, handwriting, decorative type, and brand custom fonts from a reference image, produces a font-identification report with commercial-free alternative suggestions, then transfers either the glyph shapes or the full font effect including stroke, gradient, highlight, and outer glow onto specified text in a target image while preserving position, size, and layout.","descCn":"品牌字效迁移。从参考图识别 Logo 字/泡泡字/手写字/装饰字/品牌定制字体，输出字体识别报告与商用免费替代建议，并将字形或完整字效（描边/渐变/高光/外发光）迁移替换到目标图的指定文字上，同时保留目标图的其他画面元素不变。","tools":[],"skillType":"skill","sortWeight":150,"downloads":1632,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/47709c0b-6ad0-4dd8-b192-f10da918b96b.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"video-deconstruct-analyzer","version":"0.2.5","hash":"sha256:5229c258d14718bd8fb46fc54f08db879ef52b4038d2f7a1dafcfb371a58fe89","summary":"Deconstruct AI videos with dense keyframes and Seedance prompts.","summaryZh":"高密度拆解AI视频","description":"Deep AI-video deconstruction and replica-prompt assistant. Applies when the user uploads an AI-generated video, keyframe screenshots, or a text description and asks to \"deconstruct the video\", \"analyze the video\", \"reverse-engineer the video\", \"replicate the video / camera / style / plot\", or says \"video deconstruct\". Optimized for the Hub toolchain: read video metadata first, build a canvas evidence chain from dense keyframes, output a structured Markdown deconstruction report, and generate Seedance 2.0 Chinese-language timestamp-storyboard prompts under the rule of \"prefer ≤15 seconds per prompt, use as few prompts as possible\". Primarily serves AI-video reference analysis — not general editing, subtitle translation, or long-form film criticism.","tags":["Short Drama","Commercial Ad","Animation"],"tagsCn":["短剧漫剧","商业广告","动画"],"creator":"","triggerWords":["deconstruct video","analyze video","reverse-engineer video","video replica","replicate camera","replicate style","replicate plot","dense keyframes","video-deconstruct","video deconstruct"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"视频拆解分析师","tagEn":"Short Drama / Intent Brainstorming","tagCn":"短剧漫剧 / 意图脑爆","completeTagsEn":["Short Drama / Intent Brainstorming","Commercial Ad / Intent Brainstorming","Animation / Intent Brainstorming"],"completeTagsCn":["短剧漫剧 / 意图脑爆","商业广告 / 意图脑爆","动画 / 意图脑爆"],"descEn":"Deep AI-video deconstruction with replica prompts. Reads video metadata and builds a canvas evidence chain from dense keyframes, outputs a structured Markdown deconstruction report, and reverse-engineers Seedance 2.0 Chinese timestamp storyboard prompts under the rule of preferably fifteen seconds or less per clip and as few prompts as possible. Aimed at reference analysis, not general editing or subtitle work.","descCn":"AI 视频深度拆解与复刻提示词。读取视频元信息 + 高密度关键帧建立画布证据链，输出结构化 Markdown 拆解报告，并按“每条尽量≤15 秒、提示词数量尽量少”反推 Seedance 2.0 中文时间戳分镜复刻提示词；面向参考分析，不做普通剪辑/字幕。","tools":["hub_analyse_media","hub_ffmpeg","hub_canvas_write_media_node","hub_canvas_write_text_node"],"skillType":"skill","sortWeight":160,"downloads":2004,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/c810a994-05ab-4b38-b793-59c10e04590d.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"backrooms-dreamcore","version":"0.2.5","hash":"sha256:3c529ed1c3e73f1a6225d8687a7549f4d7e548e9bbe09fd716584f0977575f55","summary":"Create personalized Backrooms dreamcore spaces from a reference image.","summaryZh":"生成专属后室空间","description":"Personalized Backrooms / dreamcore space generation skill. After the user uploads a character, pet, object, or scene image, extract spatial tendencies through 4 immersive multiple-choice questions, then combine the reference image subject with Backrooms Level 0 visual grammar to produce a shareable personalized Backrooms slice: a space dossier card, an AI image-generation brief, an optional BGM direction, and optionally continue into photoreal Backrooms image or image-to-video generation. Trigger words include: backrooms test, my backrooms, dreamcore test, my liminal space, inherent space, turn me into backrooms, backrooms-ify this scene, character backrooms-ification, scene backrooms-ification, liminal space. Not for pure Backrooms lore Q&A, generic horror monster imagery, or generic Backrooms images without personalized input.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["backrooms test","my backrooms","dreamcore test","my liminal space","inherent space","turn me into backrooms","backrooms-ify this scene","character backrooms-ification","scene backrooms-ification","liminal space"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"后室梦核","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"Transforms a user-provided character, pet, object, or scene reference image into a personalized Backrooms or dreamcore space. The skill uses four atmospheric multiple-choice questions to derive spatial tendencies along five dimensions, then produces a shareable space dossier card, an image-generation brief, an optional ambient BGM direction, and can continue into photoreal Backrooms image or image-to-video generation.","descCn":"根据用户上传的人物、角色、宠物、物体或场景图，通过四个沉浸式选择题提取空间倾向，生成专属后室 / 梦核空间档案、AI 配图需求与可选环境音乐方向，并可继续生成真实摄影感的后室图片或图生视频。适合“我的后室”“角色后室化”“场景后室化”等社媒向创作。","tools":[],"skillType":"skill","sortWeight":170,"downloads":1488,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/53e0b9a9-7d41-403d-bd4f-8aa853c60660.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"ecommerce-image","version":"0.1.11","hash":"sha256:b6542c5206a02ce1b519dd576775a3e4bea405c86ff7e337ba66ff0c8bc2fa0d","summary":"Generate platform-compliant product image set from raw photo","summaryZh":"从实拍图生成平台合规的电商组图","description":"E-commerce product image set generation. Takes a raw product photo (studio shot or phone capture) as input and produces a coherent set of platform-compliant images: hero / main image, white-background packshot, lifestyle shot, on-model shot, macro detail, multi-angle views, SKU color swatches, PDP / product-detail-page imagery, and bundle composites. Covers 14 verticals (womenswear, menswear, kidswear, footwear, bags, jewelry, beauty, consumer electronics, home & furniture, daily goods, pets, mother & baby, supplements, sports & outdoor) and follows listing specs for Amazon, Shopify, Shopee, TikTok Shop, Etsy, Instagram Shop, Pinterest, as well as Taobao, Tmall, JD, Pinduoduo, Douyin, Xiaohongshu, Kuaishou. Triggers when an actual product photo is on hand and the goal is an e-commerce listing, Amazon main image, Shopify product page, Etsy thumbnail, PDP hero, white-background catalog shot, lifestyle product scene, swatch / try-on / size-reference image, infographic listing image, or any platform-spec product photography deliverable. Not for campaign Big Idea concepts, surreal or promo videos, MV-style content, or generic AI photo enhancement without a real product reference.","tags":["E-Commerce"],"tagsCn":["电商"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"电商商品图","tagEn":"E-Commerce / Image Generation","tagCn":"电商 / 图像生成","completeTagsEn":["E-Commerce / Creative Generation"],"completeTagsCn":["电商 / 创作生成"],"descEn":"For producing platform-compliant e-commerce image sets when a real product photo is on hand. Takes a raw studio or phone shot of the product as input. Produces a coherent set of hero, white-background, lifestyle, on-model, macro detail, multi-angle, SKU swatch and PDP images for Amazon, Shopify, TikTok Shop, Taobao, Tmall and similar e-commerce platforms with automatic background and lighting optimization.","descCn":"适用于已有商品实拍照、需要批量上架或制作详情页的电商场景。输入商品原始实拍图(棚拍或手机随拍),输出主图、白底图、场景图、模特图、微距细节、多角度展示、SKU 色卡、详情页配图等成套商品图。自动优化背景与光线,适配 Amazon、Shopify、TikTok Shop、淘宝、天猫等主流电商平台规格。","tools":[],"skillType":"skill","sortWeight":180,"downloads":13328,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/e134212f-b5db-4c45-a519-066b2ae7e62d.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"character-scene-storyboard","version":"1.4.13","hash":"sha256:65a3f06a1b71fad15c3443523b7a753768baa676a123faec261697045845e91f","summary":"Comprehensive design sheet from refs and script","summaryZh":"由参考图与剧本生成综合设定图","description":"Turn 1–4 character references and a script or scene into a unified pre-production design sheet with character turnarounds, scene concept art, and adaptive storyboard panels. Not for video production.","tags":["Short Drama","Animation"],"tagsCn":["短剧漫剧","动画"],"creator":"","triggerWords":["character scene storyboard","character storyboard","design sheet","storyboard design sheet","character concept sheet","pre-production sheet"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"角色场景分镜板","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Short Drama / Planning","Animation / Planning"],"completeTagsCn":["短剧漫剧 / 计划制定","动画 / 计划制定"],"descEn":"For pre-production comprehensive design sheets combining character, scene and storyboard in one design-sheet image. Takes 1-4 character reference images plus a script or scene description as input. Produces a single design-document image with character sheets including turnaround and expression views, main scene concept art and a default 12-panel storyboard grid. Supports six visual styles for flexible art direction.","descCn":"适用于影视前期制作中将角色、场景、分镜整合到一张图上的综合设定文档。输入 1 至 4 张角色参考图与剧本或场景描述,输出单张专业设计文档图,包含角色三视图加立绘加表情集、主场景概念图、默认 12 格分镜故事板。提供 6 种视觉风格供灵活选择,满足不同美术方向的前期制作需求。","tools":[],"skillType":"skill","sortWeight":190,"downloads":4657,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/2324a97f-1b40-4f41-b0ee-30d7d459da57.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"sword-dance","version":"0.3.15","hash":"sha256:a9bddf52e05cc88b2a67e96f2a6f09f51c7edfec79e1232e22a1aa638889cbca","summary":"Generate Chinese sword dance video","summaryZh":"生成国风武侠舞剑短片","description":"MV Chinese-style animated sword dance storyboard generator. Uses Midjourney V7 to generate a stylistically unified set of wuxia images, composites character into scene via Seedance 2.0 multi-image reference, and produces the final storyboard video.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"武侠舞剑","tagEn":"Music Video / Video Generation","tagCn":"音乐视频 / 视频生成","completeTagsEn":["Short Drama / Planning"],"completeTagsCn":["短剧漫剧 / 计划制定"],"descEn":"For Chinese-style wuxia sword-dance MV shorts featuring traditional martial arts aesthetics. Takes a creative concept or character setup as input. Produces a stylistically consistent set of Midjourney V7 wuxia character images and a final composite storyboard video via Seedance 2.0 multi-image reference compositing. Maintains visual coherence across all generated assets for a unified cinematic wuxia atmosphere.","descCn":"适用于国风武侠舞剑类 MV 动画短片的制作,融合传统武术美学与现代视觉特效。输入创意概念或角色设定,输出风格统一的 Midjourney V7 武侠角色图集,经 Seedance 2.0 多图参考合成技术将角色融入场景,最终输出完整分镜成片。全流程保持视觉一致性,营造统一的电影级武侠氛围。","tools":[],"skillType":"skill","sortWeight":200,"downloads":2729,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"poster-design","version":"1.0.14","hash":"sha256:0e47057ef8647d36e831a40e98d4795c8d4d2db6c41f5eba3759bd3d776747ec","summary":"Static poster designer covering 12 subtypes","summaryZh":"覆盖12子类的静态海报设计","description":"Professional static poster design skill. Covers 12 subtypes (Xiaohongshu cover / WeChat article header / Story vertical / Movie poster / Event poster / Music festival poster / Magazine cover / Book cover / Album cover / Ad Banner / Promotional poster / Weibo Banner) + 7 anti-failure iron laws + 5 recommended composition modes (Immersive / Wrap-around / Editorial Minimal / Full Bleed / Magazine Composition) + 5-module 19-item weighted scoring system. Trigger: \"poster\", \"poster\", \"cover\", \"Xiaohongshu cover\", \"WeChat header\", \"Weibo header\", \"Story poster\", \"Moments poster\", \"banner\", \"marketing image\", \"magazine cover\", \"book cover\", \"movie poster\", \"music festival poster\", \"event poster\", \"lecture poster\", \"exhibition poster\", \"promotional poster\", \"flash sale poster\", \"album cover\", \"podcast cover\", \"ad banner\", \"landing page banner\", \"make a poster\", \"design a poster\", \"create a poster\", \"design a poster\", \"make a poster\", \"create a poster\". NOT for: animated poster short videos (use dynamic-poster) / anime character portraits (use anime-design)","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"海报设计","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Commercial Ad / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成","电商 / 创作生成"],"descEn":"For static poster design across 12 subtypes including Xiaohongshu cover, WeChat header, vertical story, film poster, music festival poster, magazine cover, book cover, album cover, ad banner, promo poster and Weibo banner formats. Takes design requirements and content brief as input. Produces a finished static poster evaluated and refined through a 5-module 19-item weighted rubric scoring system.","descCn":"适用于静态海报设计,覆盖 12 种子类型:小红书封面、公众号头图、Story 竖屏海报、电影海报、音乐节海报、杂志封面、书籍封面、专辑封面、广告横幅 banner、促销海报、微博 banner 等。输入设计需求与内容简报,输出成品静态海报。通过 5 模块 19 项加权评分体系进行质量评估与迭代打磨,确保设计产出的专业水准。","tools":[],"skillType":"skill","sortWeight":210,"downloads":2818,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/723b3a9c-86b0-4d92-ba0b-1332bc9ba8bb.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"film-style-picker","version":"0.3.10","hash":"sha256:48ef59ec4c69d4bc6854e983a4a68312bbee72c230073bb1b0e238565368a090","summary":"Pick genre and master, get reference shots","summaryZh":"选定影视类别与导演获取参考图","description":"Film style selection assistant. Triggered when a user needs to determine the visual style for AI image/video generation, find reference images, or benchmark against a specific director's or film's cinematographic aesthetics. Full workflow: Ask for style category -> Display specific styles and representative screenshots under that category -> After user locks in a style, output an AI Prompt template and optionally generate directly. During generation, automatically collects reference images from Film-Grab/Pinterest/Cosmos and other sites to anchor the style. Trigger words: find references, pick style, film style, reference director, style image, benchmark shots, image generation reference.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"影视风格选择器","tagEn":"Visual Production / Idea & Concept","tagCn":"视觉创作 / 创意构思","completeTagsEn":["Short Drama / Intent Brainstorming"],"completeTagsCn":["短剧漫剧 / 意图脑爆"],"descEn":"For locking a visual style direction before AI image or video generation by referencing established cinematic aesthetics. Takes a genre interest or director and film reference target as input. Produces a curated set of style references collected from Film-Grab, Pinterest, Cosmos and similar visual databases plus an AI prompt template ready for direct generation. Essential for maintaining visual consistency across production.","descCn":"适用于 AI 生图或生视频前确定视觉风格、找参考图、对标导演或电影镜头美学的场景,是保持制作视觉一致性的关键步骤。输入感兴趣的影视类别或导演、影片对标方向,输出该风格下的代表性参考图(自动从 Film-Grab、Pinterest、Cosmos 等视觉数据库采集)及可直接用于生成的 AI Prompt 模板。","tools":[],"skillType":"skill","sortWeight":220,"downloads":2786,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"film-shot","version":"1.0.13","hash":"sha256:026b8efcc01eb1edaf11abf1caef75f1d431aa7a3a5cdd485b9444e5ebf7d229","summary":"Cinematic shot designer with 8 styles","summaryZh":"提供8种风格的电影感分镜设计","description":"Turn a script, scene, or character brief into stills, storyboards, or character sheets. Confirm medium, apply shot language and identity locks, and review consistency. Not for posters or video.","tags":["Short Drama","Animation"],"tagsCn":["短剧漫剧","动画"],"creator":"","triggerWords":["film storyboard","cinematic still","movie shot","camera language","character sheet","expression sheet","pose library","cinematic image set"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"影视镜头与角色卡","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Short Drama / Planning","Animation / Planning"],"completeTagsCn":["短剧漫剧 / 计划制定","动画 / 计划制定"],"descEn":"For cinematic shot design, film stills and character cards with professional camera language application. Takes a script, scene description or character setup as input. Produces shots applying six camera-language dimensions including framing, angle, movement, lighting, mood and temporal, across eight visual styles such as cyberpunk, HK noir and IMAX epic. Also generates character sheets in 3, 4 and 8 views with cross-shot identity lock.","descCn":"适用于影视分镜、剧照、角色卡的专业设计,运用专业电影镜头语言。输入剧本、场景描述或角色设定,输出运用六维镜头语言(景别、机位、运镜、光影、情绪、时间)的镜头图,覆盖赛博朋克、港片黑色电影、IMAX 史诗等 8 种视觉风格。同时生成三视图、四视图和八视图角色卡,跨镜头保持角色身份一致性。","tools":[],"skillType":"skill","sortWeight":230,"downloads":3233,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"voice-clone","version":"0.4.11","hash":"sha256:22523c763aaabe18244090a24e50f9ef7e9e117266d70f089c0a83a20f29a8b6","summary":"Create an authorized reusable voice from a reference recording with consent and quality checks.","summaryZh":"根据已获授权的参考录音创建可复用音色，并完成合规和质量检查。","description":"Create a reusable cloned voice from an authorized recording with consent, source review, optional cleanup, language matching, a demo, and quality checks. Not for unauthorized imitation.","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":["voice clone","clone a voice","cloned voice","voice replica","reusable voice","voice matching"],"guidePrompt":"Please provide an authorized reference recording with clear, mostly isolated speech.","guidePromptEn":"Please provide an authorized reference recording with clear, mostly isolated speech.","displayNameZh":"音色克隆","tagEn":"Audio Content / Audio Generation","tagCn":"音频内容 / 音频生成","completeTagsEn":["Sound & Music / Creative Generation"],"completeTagsCn":["音频音乐 / 创作生成"],"descEn":"Creates a reusable cloned voice from an authorized recording. The workflow confirms consent, reviews and optionally cleans the source, matches language and demo text, checks the result, and supports safe reuse.","descCn":"根据已获授权的参考录音创建可复用克隆音色。流程包含授权确认、源录音检查与可选清理、语种和试听文本匹配、结果验收及安全复用，不用于未经授权的模仿。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":240,"downloads":2279,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"0.4.11"},{"name":"video-prompting","version":"0.1.15","hash":"sha256:f36230ac6d7b3865de2e87dc6f7ef2c53de8e36ddc290e72ddbfabf14ab65139","summary":"Craft AI video and image prompts","summaryZh":"撰写或优化AI视频图像提示词","description":"AI video/image prompt engineering expert. Triggered when users need to write or optimize prompts for AI video or image generation tools. Helps users craft high-quality generation prompts for various models (Kling, Veo3, Wan, Gemini, Midjourney, Seedance, proprietary models, etc.), incorporating model-specific best practices and formatting requirements. Trigger phrases: video prompt, image prompt, prompt engineering, how to write prompts, optimize prompt, prompt writing.","tags":["Platform Tooling"],"tagsCn":["平台工具"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"提示词专家","tagEn":"Visual Production / Prompt Engineering","tagCn":"视觉创作 / 提示词工程","completeTagsEn":["Platform Tooling"],"completeTagsCn":["平台工具"],"descEn":"For writing or optimizing prompts targeted at specific AI video and image models with model-specific tuning. Takes a creative intent or an existing prompt to optimize, plus the target model such as Kling, Veo3, Wan, Gemini, Midjourney or Seedance as input. Produces a model-tuned high-quality prompt incorporating model-specific best practices, formatting rules, parameter recommendations and prompt engineering techniques.","descCn":"适用于针对特定 AI 视频或图片模型撰写或优化提示词的场景,提供模型级别的精细调优。输入创意意图或待优化的旧 prompt、目标模型(Kling、Veo3、Wan、Gemini、Midjourney、Seedance 等),输出针对该模型调优的高质量提示词。内置各模型的最佳实践、格式要求、参数建议与提示词工程技巧。","tools":[],"skillType":"skill","sortWeight":250,"downloads":2951,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"surreal-spot","version":"0.6.17","hash":"sha256:4f377c9d7cece587279b5805b55dca556e3f39c151938ddf0e6417d2d6f52d9a","summary":"Generate surreal TVC concept videos","summaryZh":"生成超现实TVC概念视频","description":"Surreal TVC concept creator — cinematic short-form visual ads with impossible juxtaposition aesthetics. Turns product photos into TVC-grade surreal concept videos: Scene Breakdown → 9-panel storyboard grid → Seedance motion video. Professional cinematography (camera/lens/film stock per keyframe), cool desaturated palette with accent pops, lo-fi visual medium (VHS/16mm/Super8). Trigger: \"surreal spot,\" \"TVC,\" \"creative concept,\" \"surreal video,\" \"concept video,\" \"visual concept,\" \"impossible juxtaposition,\" \"concept ad,\" \"creative short film,\" \"TVC concept,\" \"surreal commercial.\" NOT for: straightforward promo (use promo-video), MV (use mv-creator), static posters.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"超现实广告","tagEn":"Advertising / Video Generation","tagCn":"广告营销 / 视频生成","completeTagsEn":["Commercial Ad / Planning"],"completeTagsCn":["商业广告 / 计划制定"],"descEn":"For TVC-grade surreal concept advertising shorts with impossible juxtaposition aesthetics. Takes a product photo and creative direction as input. Produces a scene breakdown, 9-panel storyboard and final motion video using cinematic lensing and lo-fi visual treatments including VHS, 16mm and Super8 film grain styles. Ideal for e-commerce creative shorts and brand concept advertising where visual novelty and artistic expression are priorities.","descCn":"适用于电商创意短片、TVC 概念广告等超现实视觉概念视频的制作,追求“不可能的并置”美学风格。输入产品图与创意方向,输出场景拆解、9 宫格分镜与最终动态视频,采用电影级镜头语言与 VHS、16mm、Super8 等低保真视觉媒介。特别适合需要强烈视觉冲击力与艺术表达的品牌广告场景,从概念构思到成片输出全流程覆盖。","tools":[],"skillType":"skill","sortWeight":260,"downloads":2438,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"short-drama-screenwriter","version":"0.1.13","hash":"sha256:aa8ea935f64f758e93044023d77ecb1eafc519451d684e1c57094c7341b855c3","summary":"Write full short drama scripts by topic","summaryZh":"按选题撰写完整短剧剧本","description":"Short drama screenwriting assistant. Guides the full process from topic selection to complete screenplay, with built-in hit-making methodology. Supports both domestic vertical short dramas and overseas ReelShort/DramaBox formats. Trigger words include: short drama, screenplay, screenwriter, screenwriter, short drama, micro short drama, vertical drama, episodic script, short drama creation.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"短剧编剧助手","tagEn":"Short Drama / Script Writing","tagCn":"短剧 / 剧本写作","completeTagsEn":["Short Drama / Planning"],"completeTagsCn":["短剧漫剧 / 计划制定"],"descEn":"For short drama screenwriting from topic to full episodic script covering the complete creative process. Takes a story premise or topic plus desired format such as vertical Chinese short drama or overseas ReelShort and DramaBox productions. Produces a complete episodic script with built-in viral methodology covering hooks, conflict beats, cliffhangers and audience retention techniques for maximum engagement.","descCn":"适用于微短剧从选题到完整剧本的全流程编剧,覆盖创意发想到定稿的完整创作过程。输入故事选题或题材方向、目标格式(国内竖屏短剧或海外 ReelShort、DramaBox 平台),输出分集完整剧本。内置爆款方法论,涵盖钩子设计、冲突节点编排、悬念结构设计与观众留存技巧,帮助创作者产出高完播率的短剧内容。","tools":[],"skillType":"skill","sortWeight":270,"downloads":2591,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"youtube-creator","version":"0.3.13","hash":"sha256:53aa25bb05276030352e73ab544dc34c41c85d7f5b22cc163e65fd105a0d578e","summary":"Generate YouTube script and thumbnail","summaryZh":"生成YouTube脚本与封面","description":"YouTube video creation assistant. Pipeline from topic planning to high-CTR thumbnail generation. Covers topic planning, retention-optimized scripting, and high-conversion YouTube thumbnail generation. Stops after thumbnail — does NOT generate concept images, voice, video clips, or final assembly. Trigger phrases: YouTube, youtube creator, YouTube video, make a video, YouTube script, thumbnail, YouTube SEO.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"YouTube创作","tagEn":"Documentary / Script Writing","tagCn":"纪录片 / 剧本写作","completeTagsEn":["Commercial Ad / Planning","E-Commerce / Planning"],"completeTagsCn":["商业广告 / 计划制定","电商 / 计划制定"],"descEn":"For YouTube content planning from ideation through to thumbnail design covering the pre-production phase. Takes a topic or channel direction as input. Produces topic plans with audience analysis, retention-optimized scripts with hook and payoff structure, and high-CTR thumbnail designs. Focuses on the planning and scripting stage and does not generate concept images, voiceover, video clips or final assembly.","descCn":"适用于 YouTube 视频从选题策划到封面设计的前期制作阶段,覆盖内容创作的核心规划环节。输入选题方向或频道定位,输出包含受众分析的选题规划、以 hook-payoff 结构优化留存率的脚本、高点击率封面图设计。聚焦策划与脚本阶段,不负责概念图生成、配音录制、视频片段制作与最终合成。","tools":[],"skillType":"skill","sortWeight":280,"downloads":2529,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"storyboard","version":"1.0.12","hash":"sha256:fc7a7a7db2c32b7a9d912f42b895e6bd28e200067efc7c1b3f2173bbc35f7386","summary":"Turn script into N-grid storyboards","summaryZh":"把剧本拆为N宫格分镜故事板","description":"N-grid storyboard generator. Based on a user-provided story document or description, first breaks it down into a shot-by-shot storyboard table, then calls AI to generate N-grid storyboard images. Supports reading script documents (docx/txt) for automatic shot breakdown, as well as direct story description input. Supports 2x2, 3x3, 4x4, 5x5 and other grid layouts; per-cell aspect ratio is customizable (default 16:9). Trigger words include: storyboard, story board, storyboard, N-grid, nine-grid, four-grid, grid storyboard, shot breakdown, shot breakdown, shot breakdown, draw a storyboard, help me break down shots, make a storyboard, visual script. Not for: complete documentary/educational video generation (use documentary skill), character turnaround/expression sheets (use film-tv skill), video editing compositing.","tags":["Short Drama","Animation"],"tagsCn":["短剧漫剧","动画"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"分镜板","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Short Drama / Planning","Animation / Planning"],"completeTagsCn":["短剧漫剧 / 计划制定","动画 / 计划制定"],"descEn":"For storyboard generation from script or story description with flexible grid layout options. Takes a script document such as docx or txt or a free-text story description as input. Produces a shot-by-shot breakdown table with detailed camera and action directions plus N-grid storyboard images in 2x2, 3x3, 4x4 and 5x5 layouts with customizable per-cell aspect ratios defaulting to 16:9 widescreen format.","descCn":"适用于将剧本或故事描述拆解为分镜的场景,提供灵活的宫格布局选项。输入剧本文档(docx 或 txt 格式)或自由文字故事描述,输出逐镜头分镜拆解表(含详细的镜头与动作指示)与 N 宫格故事板图。支持 2x2、3x3、4x4、5x5 四种布局,每格画面比例可自定义,默认采用 16:9 宽银幕格式。","tools":[],"skillType":"skill","sortWeight":290,"downloads":2846,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"podcast-to-content-suite","version":"0.3.1","hash":"sha256:3686ada5199a8f702b0912bcb043310812fe5146f4fefa249b062c528aa83cc4","summary":"Transform a podcast transcript into a full content marketing suite — blog, social, newsletter, show notes, audiograms, SEO.","summaryZh":"把播客转录稿转成完整内容营销套件——博客、社交、newsletter、show notes 等。","description":"Transform a podcast transcript into a comprehensive content marketing suite — SEO-optimized blog post, social media content for Twitter/X, LinkedIn, Instagram, Facebook, email newsletter, show notes, 3-5 audiogram clip scripts, and SEO elements (keywords, meta tags, schema markup). Input is a podcast transcript with timestamps; output is a complete cross-channel content ecosystem assembled from the `references/output-template.md`. Use whenever the user provides a podcast transcript, asks to repurpose a podcast episode, wants to convert audio into blog / social / newsletter content, or mentions podcast marketing or distribution.","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":["podcast to content","podcast repurposing","podcast transcript","blog from podcast","show notes","audiogram","podcast marketing","播客套件","播客二次加工","podcast distribution"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"播客套件","tagEn":"Audio Content / Post-Production","tagCn":"音频内容 / 后期制作","completeTagsEn":["Sound & Music / Post-production"],"completeTagsCn":["音频音乐 / 后期处理"],"descEn":"Podcast to Content Suite converts a podcast episode into a complete content marketing ecosystem from a single transcript. Produces an SEO-optimized blog post, social media content for Twitter/X, LinkedIn, Instagram, and Facebook, an email newsletter with subject variations, structured show notes with timestamps and resources, 3-5 audiogram clip scripts, and SEO elements including keywords, meta tags, and schema markup. Uses two reference files — component-specs.md for per-asset requirements and output-template.md for the final assembly format.","descCn":"播客转内容套件从一份播客转录稿产出完整的内容营销生态。产物包括 SEO 优化的博客、Twitter/X + LinkedIn + Instagram + Facebook 的社交内容、含主题行多版本的邮件 newsletter、带时间戳和资源的结构化 show notes、3-5 个音频卡片脚本，以及关键词、meta 标签、schema markup 的 SEO 元素。使用两份 reference 文件——component-specs.md 定义每类产物的要求，output-template.md 定义最终组装格式。","tools":["hub_read","hub_write","hub_save_file_to_session","question"],"skillType":"skill","sortWeight":300,"downloads":2245,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"native-ad-mockup","version":"0.3.1","hash":"sha256:0383c4f2da6e502aa4aaafeff8eedd35cf47f01299785c8dc9515bdbb552c1cf","summary":"Create native soft-ad images inside realistic phone mock frames for products and social-commerce formats.","summaryZh":"为产品生成手机原生软广样机。","description":"Generate native-looking soft-ad static images inside realistic mobile mock frames using the user's product, product images, product page, or typed product facts. Supports iMessage/DM, WeChat-style chat, Xiaohongshu notes, comment Q&A, product cards, checkout/cart, low-battery alerts, reminders, weather/body-status apps, workplace chat, ingredient cards, scarcity cards, testimonials, and comparison ads. It asks whether product images are ready, asks which native-ad format to use, then produces prompts and images with phone-frame, inner-screen proportion, system-font, top-whitespace, and text-readability rules. Use whenever the user wants 手机 mock 外框、原生广告样机、软广图片、iMessage 广告、微信聊天推荐、小红书电商图、评论区问答、商品卡、低电量弹窗、App 截图风广告, or a set of native UI static ads for a product.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["native-ad-mockup","原生广告样机","手机mock","软广图片","iMessage广告","微信聊天推荐","小红书电商图","商品卡"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"原生广告样机","tagEn":"Advertising / Idea & Concept","tagCn":"广告营销 / 创意构思","completeTagsEn":["Commercial Ad / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成"],"descEn":"Native Ad Mockup creates product soft-ad images that look like real mobile screenshots inside a slim phone frame. It first asks whether product images are ready and which native format to use, then generates prompts and images for iMessage/DM, WeChat-style chat, Xiaohongshu notes, comment Q&A, product cards, checkout pages, alerts, reminders, weather apps, testimonials, and more. It locks inner-screen proportion, system-font feel, top whitespace, and readable text.","descCn":"原生广告样机用于把产品做成像真实手机截图一样的软广图。它会先询问是否已有产品图、想做哪类原生软广，再生成提示词和图片。支持 iMessage、微信聊天、小红书种草、评论区问答、商品卡、结账页、低电量弹窗、提醒事项、天气状态、评价卡等，并内置手机外框、内屏比例、系统字体、顶部留白和可读性规则。","tools":[],"skillType":"skill","sortWeight":310,"downloads":2255,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"film-assets","version":"0.4.11","hash":"sha256:a66f0cb0d6ce943e04cbb4ae43676e02f9438cc1e795c44d7eceb493f9bc97e8","summary":"Bulk multi-view asset sheets generation","summaryZh":"批量生成多视角资产参考图","description":"Create film assets from references or text. Confirm type, output form, variant count, background, and style; deliver character sheets, prop multi-views, or scene concepts. Not for video or audio.","tags":["Short Drama","Animation"],"tagsCn":["短剧漫剧","动画"],"creator":"","triggerWords":["-","character","card"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"影视资产","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Short Drama / Creative Generation","Animation / Creative Generation"],"completeTagsCn":["短剧漫剧 / 创作生成","动画 / 创作生成"],"descEn":"For batch generation of pre-production reference assets covering characters, props and environments. Takes a target asset description such as character, prop or scene as input. Produces character cards with three, four and eight-view turnarounds, expression sheets, pose sheets, prop multi-view sheets and scene concept art with lighting, season and weather variants. Automatically determines optimal aspect ratio and layout composition.","descCn":"适用于影视前期批量生产参考资产,覆盖角色、道具与环境三大类别。输入目标资产(角色、道具或场景)描述,输出角色三视图、四视图、八视图转身图、表情包合集、姿势组图、道具多视图展示图、场景概念图(含光影变体、季节变体、天气变体)。自动判断最优画幅比例与排版构图方式。","tools":[],"skillType":"skill","sortWeight":320,"downloads":2603,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"voice-design","version":"2.0.8","hash":"sha256:4165a2e2a7ed0cc46fa2f2a2fdc0b6ae57aee983d02aba3566e6b779ca24a517","summary":"Design and preview custom voices from text","summaryZh":"用文字描述设计并试听自定义音色","description":"Design custom AI voices by describing voice characteristics in natural language. Provide a text description of the desired voice and a preview text to hear a sample. Returns a reusable voice_id for speech synthesis. IMPORTANT: Designed voices are experimental and generally lower quality than the built-in voice library (get_voice_id). Always recommend users try the official voice library first — only use this skill when the user explicitly asks to design/create a custom voice, or when no suitable voice exists in the library. Trigger: design a voice, create custom voice, 设计声音, 自定义音色, 设计一个XX声音, 帮我设计一个声音。","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":["声音设计","设计声音","自定义声音","voice design","custom voice","design voice"],"guidePrompt":"帮我设计一个温暖沉稳的女性旁白声音","guidePromptEn":"Help me design a warm and calm female narrator voice","displayNameZh":"声音设计","tagEn":"Audio Content / Audio Generation","tagCn":"音频内容 / 音频生成","completeTagsEn":["Sound & Music / Creative Generation"],"completeTagsCn":["音频音乐 / 创作生成"],"descEn":"For designing custom AI voices when no suitable voice exists in the standard library and a unique voice identity is required. Takes a natural-language description of the desired voice characteristics plus a preview text as input. Produces a sample audio for previewing and a reusable voice_id for TTS integration. This is an experimental feature and quality may be lower than the built-in voice library.","descCn":"适用于官方音色库找不到合适声音、需要自定义设计独特音色身份时使用。输入对声音特征的自然语言描述(如年龄、性别、情感色彩、语速等)与试听文本,输出试听样本音频与可复用的 voice_id 供 TTS 集成使用。此功能为实验性质,品质通常低于内置音色库,建议优先使用 get_voice_id 检索现有音色。","tools":["hub_design_voice"],"skillType":"skill","sortWeight":330,"downloads":2433,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"multi-shot","version":"1.0.10","hash":"sha256:f91ef19624bdcde4afd1c683b8c9cf20d9ca83cf834360512827f4ecad804bbe","summary":"Re-render image from a new camera angle","summaryZh":"用新相机视角重新渲染图片","description":"Re-render one image from a new camera angle using rotation, tilt, and zoom while preserving identity, geometry, materials, and style. Use for alternate views, not changes to the depicted event.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["multi-angle","change angle","change shot","multi shot","change angle","new camera angle"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"多角度","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For creators who need an alternate camera view of an existing image without redesigning its subject or event. Provide one source image, choose a new horizontal angle, vertical angle, and framing distance, then identify the face, product shape, room layout, text, or other details that must remain stable. The workflow analyzes visible geometry, explains which hidden surfaces require inference, confirms risky viewpoint changes, produces the alternate view, and checks identity, proportions, lighting, scene layout, and typography. It delivers one re-rendered image plus a concise camera-view note. It is best for plausible viewpoint changes, not inventing unseen product details or changing the depicted action.","descCn":"面向需要为已有图片补充新机位、但不希望重做主体或事件的创作者。用户提供一张原图，选择水平角度、俯仰方向和景别距离，并说明人脸、产品外形、房间布局、文字或其他必须保留的细节。流程先分析原图可见结构，指出新视角中需要推测的遮挡面，再确认风险较高的机位变化，生成新视角并检查身份、比例、光线、场景关系与文字一致性。最终交付一张重渲染图片和简短机位说明，适合合理换视角，不用于凭空补全精密产品结构或改变原有动作。","tools":[],"skillType":"skill","sortWeight":340,"downloads":2310,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"content-strategy","version":"0.3.1","hash":"sha256:c9fbef090a232d3174986d5c4e80c523198065e0aed6006685a9676e6808c92b","summary":"Content strategist that decides what to write via searchable/shareable framework, buyer-stage mapping, and 4-factor scoring.","summaryZh":"内容策略师帮你选题：搜索传播分类、购买阶段、四维打分。","description":"Content strategist that helps decide what content to produce (not just write it). Gathers business/customer/competitive context, applies the searchable vs shareable framework, maps ideas to buyer stages, scores by customer impact + content-market fit + search potential + resources, and outputs content pillars, priority topics, and a topic cluster map.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["内容策略","内容支柱","话题簇","编辑日历","内容路线图","content strategy","content pillars","topic cluster","editorial calendar","content roadmap","blog topics","content ideas"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"内容策略","tagEn":"Advertising / Idea & Concept","tagCn":"广告营销 / 创意构思","completeTagsEn":["Commercial Ad / Intent Brainstorming","E-Commerce / Intent Brainstorming"],"completeTagsCn":["商业广告 / 意图脑爆","电商 / 意图脑爆"],"descEn":"Content Strategy helps decide what content to produce, not just write it. Gathers business, customer research, current state, and competitive context, then applies the searchable vs shareable framework across use-case, hub/spoke, thought-leadership, data-driven, and case-study types. Maps ideas to awareness / consideration / decision / implementation stages, scores by customer impact (40%) + content-market fit (30%) + search potential (20%) + resources (10%), and outputs content pillars, priority topics, and a topic cluster map.","descCn":"Content Strategy 帮用户决定要做什么内容，而不只是写。搜集商业背景、客户研究、现状与竞品格局，把 searchable vs shareable 框架应用到 use-case、hub/spoke、思想领导力、数据驱动、案例研究等类型上。把 idea 映射到 awareness / consideration / decision / implementation 四个购买阶段，按客户影响 (40%) + 内容市场契合 (30%) + 搜索潜力 (20%) + 资源 (10%) 打分，输出内容支柱、优先话题与话题簇地图。","tools":["question","read","hub_read","hub_write","hub_save_file_to_session"],"skillType":"skill","sortWeight":350,"downloads":2240,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"voiceover-direction","version":"0.3.1","hash":"sha256:da8e4af72dec52eb54f9a10e12b8c11114c98e5fde7e8663126435733122b7a5","summary":"Direct voice talent from brief and speakable script through casting, live-session, and pickup feedback — intention-based, not inflection-based.","summaryZh":"配音导演全流程：brief、脚本、选角、现场、补录反馈——讲意图不讲抑扬。","description":"Direct voice talent to deliver performances that match brand vision — from writing the VO brief and speakable script, through casting the right voice, to running the session live and giving actionable pickup feedback. Input is a project (SaaS explainer, commercial, e-learning, audiobook, ad) plus brand and audience context; output is a VO brief, a speakability-checked script, audition criteria, live-session direction notes with intention-based cues, and structured pickup notes. Use whenever the user wants to hire and brief voice talent, prep a live session, redirect a bad take without insulting the artist, cast the right voice for a brand, rewrite a script that isn't easy to speak, or give clear pickup notes on a delivered read.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":["配音","配音导演","VO brief","配音脚本","补录反馈","选角","voiceover","voice direction","VO brief","speakable script","casting voice","pickup notes","voice talent"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"配音导演","tagEn":"Audio Content / Post-Production","tagCn":"音频内容 / 后期制作","completeTagsEn":["Short Drama / Planning"],"completeTagsCn":["短剧漫剧 / 计划制定"],"descEn":"Voiceover Direction covers the full workflow of directing voice talent: writing a specific VO brief with sound-alike and anti-references, rewriting scripts to be speakable (15-word sentences, contractions, breathing room, phonetic spellings), casting by acoustic and performance qualities, running the live session with intention-based cues rather than inflection commands, and giving structured pickup feedback that names timestamp, issue, desired change, and story reason. Includes templates, checklists, and worked SaaS-explainer and redirect examples.","descCn":"配音导演 skill 覆盖导演配音演员的完整流程：写有 sound-alike 和 anti-reference 的具体 VO brief；把脚本改到好念（15 词短句、连读、呼吸空间、音标标注复杂词）；按声学与表演特质选角；用「意图」而非「抑扬」命令来跑现场 session；给带时间戳、当前问题、期望改动和故事原因的结构化补录反馈。附模板、checklist 以及 SaaS 解说和现场掰回的实战示例。","tools":["question","hub_read","hub_write","hub_save_file_to_session","hub_audio_generation"],"skillType":"skill","sortWeight":360,"downloads":2152,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"n-storyboard","version":"1.0.12","hash":"sha256:780ff564ada9f331eea6020829c514eb309558a34788c1bb16435930ea3f9589","summary":"Multi-panel storyboard from a story","summaryZh":"由故事生成多宫格分镜整图","description":"Generates a multi-panel storyboard composite image based on story text and grid settings (2x2/3x3/4x4) + per-cell aspect ratio + optional reference images.","tags":["Short Drama","Animation"],"tagsCn":["短剧漫剧","动画"],"creator":"","triggerWords":["multi-panel storyboard","n-panel storyboard","story panel grid","storyboard","story panel grid","multi panel storyboard"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"多宫格分镜","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Short Drama / Planning","Animation / Planning"],"completeTagsCn":["短剧漫剧 / 计划制定","动画 / 计划制定"],"descEn":"For producing a single composite N-panel storyboard image from a story in one unified visual output. Takes a story text plus grid setting such as 2x2, 3x3 or 4x4 layout, per-cell aspect ratio and an optional reference image as input. Produces one stitched multi-panel storyboard image rather than separate individual shots, ideal for quick visual storytelling previews and pitch presentations.","descCn":"适用于将故事直接生成一张 N 宫格分镜整图的场景,以单张统一视觉输出呈现完整故事。输入故事文本、宫格设置(2x2、3x3 或 4x4 布局)、单格画面比例与可选参考图,输出一张拼合好的多宫格分镜整图(非分散的独立单镜)。特别适合快速视觉叙事预览、项目提案展示和创意沟通场景。","tools":[],"skillType":"skill","sortWeight":370,"downloads":2176,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"ad-creative","version":"0.3.2","hash":"sha256:e08278f7e287cee9f9ffcb4cf5461cb1d2ceb0c88964dc94f6a9a664b30c41f8","summary":"Generate, scale, and iterate paid ad copy across Google, Meta, LinkedIn, TikTok, X — angle-organized variations validated against platform limits.","summaryZh":"跨 Google/Meta/LinkedIn/TikTok/X 大规模生成、迭代广告文案，按角度组织并校验字符上限。","description":"Generate, iterate, and scale ad creative — headlines, descriptions, primary text, or full ad variations — for any paid advertising platform (Google Ads RSA, Meta, LinkedIn, TikTok, Twitter/X). Input is product / audience / platform context, or existing ads plus performance data (CSV / paste / API); output is a bank of angle-organized variations that pass each platform's character limits, plus an iteration report when performance data is provided. Use whenever the user wants to write ad copy at scale, generate RSA headlines, produce Facebook ad copy or LinkedIn ad text, request \"more ad variations\" or \"ad copy variations\", set up creative testing, or iterate ads from campaign performance data.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["广告文案","RSA标题","广告变体","广告创意","投放文案","ad creative","ad copy","RSA headlines","ad variations","Meta ads","LinkedIn ads","TikTok ads","creative testing"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"广告创意","tagEn":"Advertising / Idea & Concept","tagCn":"广告营销 / 创意构思","completeTagsEn":["Commercial Ad / Planning"],"completeTagsCn":["商业广告 / 计划制定"],"descEn":"Ad Creative is a performance-marketer skill that generates and iterates ad copy at scale for Google Ads RSA, Meta, LinkedIn, TikTok, and Twitter/X. Supports two modes: from-scratch generation with 8 angle categories, and data-driven iteration that analyzes winners, retires losers, and doubles down on winning themes. Every variation is validated against per-platform character limits, delivered organized by angle with counts or bulk CSV for direct upload. Includes a batch generation waves workflow for 100+ variations per cycle.","descCn":"广告创意 skill 是一套面向表现广告的 copy 生成工具，覆盖 Google Ads RSA、Meta、LinkedIn、TikTok、Twitter/X 五个主流平台。支持两种模式：基于 8 类角度的从零生成，和基于投放数据的迭代（识别赢家 pattern、下线输家、加码赢家）。每一条变体都按平台字符上限校验，按角度分组交付，附字符数或可直接上传的批量 CSV。内含分波次批量工作流，一轮产出 100+ 变体。","tools":["question","hub_read","hub_write","hub_save_file_to_session"],"skillType":"skill","sortWeight":380,"downloads":2119,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"anime-style-forge","version":"0.2.11","hash":"sha256:f0a08b120052a5fa4ffcc56229c7930de03a6540b1f2bca63c0f26becc88ef30","summary":"Forge anime characters in target styles","summaryZh":"把角色锻造为目标二次元画风","description":"Specialized in anime/2D/character stylization for image generation and conversion. Covers Japanese, Chinese, Korean, and Western art style families. Uses provenance analysis to trace reference images' style DNA, performs a 10-dimension analysis → 3-dimension collapse to precisely lock the style's essence, then matches the optimal tool and prompt approach for generation. Trigger on: \"anime-ify\", \"2D style\", \"convert to anime\", \"cel-shading\", \"ghibli style\", \"Korean watercolor\", \"fantasy 3D\", \"chibi\", \"Japanese anime style\", \"style conversion\", \"manga style\", \"character illustration\", \"anime style\", \"webtoon style\", or any request involving converting content into a specific anime/2D art style. Key distinction: User requests generation or conversion to a specific anime/2D art style. Do NOT trigger for: photorealistic photography style, pure logo design, general image editing (crop/background removal etc.).","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"二次元风格锻造","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For anime style transfer and character stylization across Japanese, Donghua, Manhwa and Western anime systems. Takes a reference image and target style direction as input. Produces stylistically reforged images via 10-dimension style decomposition covering line weight, color palette, shading method and more, collapsed into 3-axis Form, Aesthetic and Mood positioning, with matched generation tooling per style genome.","descCn":"适用于二次元及动漫角色的风格化生成与画风转换,覆盖日系、国漫、韩系、欧美等画风体系。输入参考图与目标风格方向,输出复刻该风格的新图。通过 10 维拆解(线条粗细、色彩方案、着色方式等)→ Form、Aesthetic、Mood 三轴坍缩定位画风基因,自动匹配最优生成工具与 prompt 写法,确保风格转换的精准度。","tools":[],"skillType":"skill","sortWeight":390,"downloads":1976,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"anime-design","version":"1.0.11","hash":"sha256:417ec2f695bab02774a6559b58e0b7fd2580bd54978eb0a99ae46d07b56163a8","summary":"14-substyle anime expert with character lock","summaryZh":"14种二次元画风且锁定角色一致","description":"Professional anime/2D art style generation skill. Covers 14 sub-styles (modern Japanese anime/moe, retro Japanese cel-shading, Japanese shonen, Japanese shojo, Ghibli, Makoto Shinkai, Chinese xianxia/ink wash, modern Chinese anime, Chinese 3D fantasy, Korean webtoon, Korean impasto, Western cartoon, chibi/moe, 2D cyberpunk) + 5 anti-failure iron laws + cross-style shared rules (character lock / facial proportion spec / stroke consistency / universal negative). Core capabilities: precise style targeting, consistent character identity, cross-style conversion. Trigger: \"anime\", \"2D art\", \"manga\", \"illustration\", \"Japanese anime\", \"Chinese anime\", \"Korean webtoon\", \"webtoon\", \"Western cartoon\", \"ghibli\", \"shinkai\", \"ufotable\", \"trigger style\", \"cel-shading\", \"impasto\", \"chibi\", \"moe\", \"catgirl\", \"Chinese 3D fantasy\", \"xianxia\", \"ink wash\", \"hanfu character\", \"cyberpunk anime\", \"draw an anime character\", \"make an anime avatar\", \"anime character\", \"anime style\". NOT for: photorealistic (use image agent default) / static posters (use poster-design)","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"二次元设计","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For anime-style image generation across 14 substyles including modern moe, Showa cel shading, Ghibli watercolor, Shinkai lighting, Donghua wuxia, Manhwa webtoon, US cartoon, chibi and cyberpunk anime aesthetics. Takes character description, style preference and optional reference image as input. Produces anime images with character lock for identity consistency, facial proportion rules and style coherence across multiple generations.","descCn":"适用于二次元及动漫风格的图片生成,覆盖 14 种子风格:日漫现代萌系、昭和赛璐璐、吉卜力水彩、新海诚光影、国漫仙侠水墨、韩漫 webtoon、美漫卡通、Q 版、二次元赛博朋克等。输入角色描述、画风偏好与可选参考图,输出锁定画风、保持角色身份一致的二次元图像,遵循面部比例规则,确保多次生成间的风格连贯性。","tools":[],"skillType":"skill","sortWeight":400,"downloads":2005,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"relight","version":"1.0.18","hash":"sha256:80a2814b27381418c9eef20c827e16502e36e8cc51a6cf75169d28a8dd366471","summary":"Smart relight with direction, intensity and color","summaryZh":"智能调整光线方向、强度与色彩","description":"After uploading an image, control the light direction / intensity / color via a 3D lighting sphere panel, select background mode and atmosphere effects, and relight the scene with one click. Provides 20 style presets and 25 atmosphere effects, and also supports free editing of up to 3 light sources (type / angle / color temperature or HEX).","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["relight","change lighting","lighting","relight","change lighting","re-light"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"光影工作室","tagEn":"Visual Production / Image Generation","tagCn":"视觉创作 / 图像生成","completeTagsEn":["Commercial Ad / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成","电商 / 创作生成"],"descEn":"For repainting lighting on an existing image with precise directional and color control. Takes one source image plus lighting controls via a 3D light-sphere panel including direction, intensity and color temperature, plus background mode and ambient effect choices as input. Produces a relit image with 20 style presets, 25 ambient effects and up to 3 free-edit lights configurable by type, angle and color temp or HEX.","descCn":"适用于已有图片的重新打光,提供精确的光线方向与色彩控制。输入一张原图,通过 3D 灯光球面板控制光线方向、强度与色温,选择背景模式与氛围特效,输出重新照亮后的新图。提供 20 种风格预设、25 种氛围特效,支持自由编辑最多 3 盏灯源,每盏可配置灯光类型、入射角度与色温值或十六进制色值。","tools":[],"skillType":"skill","sortWeight":410,"downloads":1791,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"writing-fragments","version":"0.3.1","hash":"sha256:d247d7534e0cf1948c30c2d2dfc64e8e3319292ab4e4f22a2be594f2547380d7","summary":"Grill the user for heterogeneous writing fragments and append them to a single markdown file as raw material for a future article.","summaryZh":"拷问式访谈挖出写作碎片,追加到 markdown 文件作为成文原料。","description":"Run a grilling interview session that mines the user for fragments — heterogeneous nuggets of writing (claims, vignettes, sharp sentences, half-thoughts, quotes) — and appends them to a single markdown file as raw material for a future article. Input is the writing topic and a save path; output is a growing fragments document with each nugget separated by `---`, no imposed structure or outline. Use whenever the user wants to develop ideas before imposing structure, mentions \"fragments\", \"ideate\", \"raw material\", or wants to interview themselves out loud on a writing topic.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["writing fragments","写作碎片","ideate","raw material","article fragments","写作素材","self-interview","novelist diary","brainstorm writing","grilling interview"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"写作碎片","tagEn":"Platform Tooling","tagCn":"平台工具","completeTagsEn":["Creative & Experimental / Intent Brainstorming"],"completeTagsCn":["创意实验 / 意图脑爆"],"descEn":"Writing Fragments runs a relentless interview session that mines the user for fragments — sharp sentences, claims, vignettes, half-thoughts, quotes, observations. Each nugget is appended to a single markdown file separated by horizontal rules, with a working title H1 at the top and no imposed structure, tags, or outline. Re-reads the file before every write to preserve user edits. The novelist's diary as a model for writing raw material.","descCn":"写作碎片跑一场拷问式访谈,持续追问用户的写作话题,把冒出来的犀利句子、观点、场景、半成形想法、引言、观察全部挖成 fragments,追加到同一个 markdown 文件,碎片之间用水平线分隔,顶部只放一个工作标题 H1,不加大纲、tag、结构。每次写入前都重新读文件以保留用户手动编辑。对标小说家日记的原料积累方式,让想法在被结构约束之前先自由展开。","tools":["hub_read","hub_write","hub_edit","question","hub_save_file_to_session"],"skillType":"skill","sortWeight":420,"downloads":1675,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"3d-animation-short-generator","version":"1.1.17","hash":"sha256:f3c9653a6099b358f1b995e470df369aea6000abbc0974f8aefa120dd153c4d1","summary":"Create a unified 3D animated short from a story idea, covering characters, scenes, shots, storyboards, video clips, and final assembly.","summaryZh":"根据故事创意完成角色场景设定、镜头规划、分镜生成与视频合成，输出风格统一的3D动画短片。","description":"Create 3D animated shorts from one-line ideas through briefs, cards, shot tables, storyboards, clips, BGM, and final assembly. Use for stylized story-to-video workflows with audio-safe lip-sync.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["3d animation short","animated short","pixar-style short","story-to-video","cartoon short","lip-sync safe"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"3D动画短片","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Planning","Animation / Creative Generation","Animation / Post-production"],"completeTagsCn":["动画 / 计划制定","动画 / 创作生成","动画 / 后期处理"],"descEn":"Turns a short story idea into a stylized 3D animated short. It builds the brief, outline, character and environment cards, continuity-safe shot table, text storyboard, single-shot clips, music, assembly, and final quality review. MiniMax-H3 is the default video model; another model is used only after the user explicitly selects it and its capabilities are checked. Best for coherent cartoon shorts, not single images or unrelated clips.","descCn":"面向希望把故事创意制作成完整风格化3D动画短片的创作者。用户提供故事前提、时长、画幅、基调和音频模式后，Skill 依次完成项目简报、故事大纲、角色卡、场景卡、连续性镜头表、文本分镜、单镜头视频、配乐、成片合成与质量复查。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力再执行，失败会按原因重试并切换可用模型。适合动画短片，不适合单图或无关联镜头，强调角色统一与镜头衔接。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":430,"downloads":4799,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/96493a75-aa47-4163-9bf0-14ba3e0ee3fe.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"1.1.17"},{"name":"brand-promo-video-generator","version":"0.1.15","hash":"sha256:442269f25537e8a7d8b657298bc613e9004d3333929a5cbcbaa08759c2fd0012","summary":"Turn verified brand assets and campaign goals into a polished promotional short video.","summaryZh":"基于品牌素材与推广目标，完成事实核验、创意方向、分镜和音画合成，输出可展示的品牌宣传短片。","description":"Turn authorized assets, links, or a campaign concept into a brand promo. Verify facts and sources, plan beats, produce and assemble media. Not for false claims, unauthorized imitation, or long films.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["brand promo video","product launch video","website promo","app promo video","social campaign video","promotional reel"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"品牌宣传短片生成器","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Planning","Commercial Ad / Creative Generation","Commercial Ad / Post-production"],"completeTagsCn":["商业广告 / 计划制定","商业广告 / 创作生成","商业广告 / 后期处理"],"descEn":"For marketers and creators producing promotional content for brands, products, websites, apps, shops, or personal projects. Users provide logos, product images, interface screenshots, official links, or other verifiable assets and confirm duration, aspect ratio, audience, and campaign focus. The Skill organizes brand facts and asset provenance, selects a narrative direction, plans precise beats and shots, generates needed imagery, video, voiceover, or music, and completes assembly and pre-delivery review. It outputs a promotional short that highlights product capabilities, use cases, and a call to action. Best for launches, website showcases, and social promotion; not for imitating real brand marks without authorized assets, inventing product claims, or producing long-form narrative films.","descCn":"面向需要为品牌、产品、网站、App、小店或个人项目制作宣传内容的运营与创作者。用户需提供 LOGO、产品图、界面截图、官网链接或其他可核验素材，并确认时长、画幅、受众和推广重点。Skill 会整理品牌事实与素材来源，选择叙事方向，规划精确节拍和镜头，生成所需图像、视频、旁白或音乐，并完成合成与交付前检查。最终输出一条突出产品功能、使用场景和行动号召的品牌宣传短片。适用于新品发布、官网展示和社交媒体推广，不适用于缺少授权素材时仿造真实品牌标识、虚构产品功能或制作长篇剧情影片。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":440,"downloads":4167,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620863948845933-1786620863940.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"0.1.15"},{"name":"minimalist-product-ad-generator","version":"1.0.3","hash":"sha256:7617b93c425e06c5fe93874be9e4a3bb982a84a25bd5386665562707d54683c4","summary":"Turn product images and ad requirements into minimalist product ad shorts","summaryZh":"基于产品图片和广告需求，完成卖点提炼、英文文案与节拍分镜设计，输出极简产品广告短片，适用于电商推广与新品发布。","description":"Turn product images into a minimalist ad with preserved identity, concise English copy, independent anchors, a beat storyboard, and music replacement. Not for talking-head ads or screen demos.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["minimalist product ad","premium product ad","minimalist product film","Apple-style product ad","ecommerce product film","product launch ad"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"极简产品广告生成器","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Planning","Commercial Ad / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["商业广告 / 计划制定","商业广告 / 创作生成","电商 / 创作生成"],"descEn":"Upload a product image and create a polished Apple-style ad short for e-commerce products, small-brand launches, food, toys, gadgets, accessories, and other physical goods. The Skill helps confirm format, extract selling points, write concise English copy, create product anchors, plan beat-synced typography, and generate a clean product film with premium camera language.","descCn":"面向电商卖家、小品牌主理人和独立创作者，用于把食品、潮玩、数码配件等实体产品图片制作成极简高级广告短片。用户提供产品图、款式信息和广告需求后，Skill 会确认时长、画幅与文案方式，提炼卖点并生成简洁英文广告语，制作三张独立产品锚定图，规划文字卡点和精确节拍分镜，再默认使用 H3 生成具有留白、细节特写与高级镜头语言的产品片；如果原生音乐不理想，可按需替换独立配乐。最终交付适用于电商推广和新品发布的成片，不用于口播广告、普通剪辑或复杂屏幕演示。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":450,"downloads":3324,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620876704323439-1786620876698.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"1.0.3"},{"name":"music-video-subtitle-generator","version":"0.6.11","hash":"sha256:c6aa5bd4b222653954a2be8deb40737374227e490fc7a6a563061917e0eb1814","summary":"Create beat-synced MV prompts and lyric typography plans from music and lyrics.","summaryZh":"基于音乐和歌词完成节奏拆解、镜头设计与动态字幕方案，适用于人工智能音乐视频创作。","description":"Create a stylized AI music video or emotional short with lyric typography from music, lyrics, references, characters, mood, and platform needs. Analyze beat and vocals, separate visual and text references, design beat-reactive type, break long works into connected shots, generate H3 clips by default, and assemble them against one master audio track. Deliver concepts, shot prompts, lyric plans, and a finished edit. Not for ordinary caption cleanup, licensed-IP copying, or fully manual editing.","tags":["Sound & Music","Creative & Experimental"],"tagsCn":["音频音乐","创意实验"],"creator":"","triggerWords":["MV","music video","lyric typography","on-screen text","prompt audit","Trap MV","Gospel hip-hop","Dark-pop","Cyber-grunge"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"音乐MV动态字幕生成器","tagEn":"Sound & Music","tagCn":"音频音乐","completeTagsEn":["Sound & Music / Planning","Sound & Music / Creative Generation","Creative & Experimental / Creative Generation"],"completeTagsCn":["音频音乐 / 计划制定","音频音乐 / 创作生成","创意实验 / 创作生成"],"descEn":"For musicians, video creators, and social-media editors producing AI music videos or emotional short films with lyric typography. Users provide music, lyrics, references, characters, typography direction, mood, or target platform. The Skill analyzes beat and vocal timing, separates character, scene, and text references, designs beat-reactive spatial typography, decomposes long works into connected shots, audits prompts, and routes generation for H3 or other video tools. It outputs MV concepts, shot prompts, lyric text plans, and stitching guidance. Best for stylized MVs and subtitle-driven music visuals, not ordinary caption cleanup, licensed IP copying, or fully manual post-production editing.","descCn":"面向音乐人、视频创作者和社交媒体剪辑者，用于制作带歌词贴字的 AI MV 或情绪短片。用户提供音乐、歌词、参考图、角色、字体方向、情绪意图或发布平台。Skill 会分析节拍与人声时序，区分人物、场景和文字参考，设计随节奏变化的空间字幕，拆解长视频为可衔接镜头，审查提示词并路由 H3 等视频生成工具。最终输出 MV 概念、分镜提示词、歌词文字方案和拼接建议。适合风格化音乐视觉，不适合普通字幕校对、照搬已有 IP 或完全手工后期剪辑。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":460,"downloads":3068,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620878530922185-1786620878522.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"0.6.11"},{"name":"paper-collage-explainer-generator","version":"0.3.10","hash":"sha256:7783181fd477dfd218fc48236369f862feb4d8ea8b1a10d4e339ac22610466d6","summary":"Turn scripts, story beats, and abstract ideas into approval-ready paper-collage explainer animations with tactile sound design.","summaryZh":"将文案、故事节点或抽象概念转成可确认的纸拼贴讲解动画，并保留触感音效。","description":"Turn narration, opinions, or abstract ideas into paper-collage stills and stop-motion clips with tactile SFX and optional assembly. Not for presenter ads, editable layers, or prompt-only work.","tags":["Education","Animation","Creative & Experimental"],"tagsCn":["教育","动画","创意实验"],"creator":"","triggerWords":["paper collage explainer","paper-collage animation","halftone collage","collage explainer","paper collage","stop-motion collage","collage animation"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"纸拼贴讲解动画","tagEn":"Education","tagCn":"教育","completeTagsEn":["Education / Creative Generation","Animation / Creative Generation","Creative & Experimental / Creative Generation"],"completeTagsCn":["教育 / 创作生成","动画 / 创作生成","创意实验 / 创作生成"],"descEn":"For creators, educators, and social-video editors who need a tactile visual language for narration lines, knowledge points, opinions, or abstract topics. Users provide short copy, story beats, or a core concept, with optional aspect ratio, duration, palette, and audio needs. The Skill extracts the meaning, proposes visual metaphors, prepares an approval-ready production plan, generates and confirms halftone collage still frames, then creates stop-motion clips with paper slides, pops, taps, presses, and rustles. It delivers cohesive editorial paper-collage animation assets or an assembled explainer video, and it is not intended for realistic presenter ads, precisely editable layer files, complex typography systems, or prompt-only handoff.","descCn":"面向内容创作者、教育讲解者和社交视频编辑，用触感纸拼贴语言表现口播句、知识点、观点或抽象主题。用户提供短文案、故事节点或核心概念，也可补充画幅、时长、色调和音频需求。Skill 会提炼含义与视觉隐喻，先生成可确认的制作方案，再产出并确认半调纸拼贴静帧，随后制作包含纸片滑入、弹入、轻敲、压平和摩擦声的停格动画片段。最终交付统一的纸拼贴动画素材或完整讲解视频，不面向真人广告、精确可编辑图层、复杂文字系统或仅提示词交付。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":470,"downloads":2966,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620880431282617-1786620880423.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"0.3.10"},{"name":"fpv-tour-video-generator","version":"1.0.1","hash":"sha256:faa61ba66de2c1c15797e2c8fe5d466be883a1f6f4ccef634b5b401abdd19c8b","summary":"Turns uploaded characters and environments into first-person FPV tours by confirming duration and ratio, then choosing route geometry that fits the scene.","summaryZh":"先确认时长和画幅，再按场景空间决定路线几何，生成连续沉浸的第一人称 FPV 一镜到底穿越视频。","description":"Create scene-first FPV one-take flythroughs from characters and environments, with the character as route guide. Confirm duration and ratio, keep character identity separate from scene style, and plan dives, banks, gaps, near misses, and parallax. Use a paper airplane if no character is supplied. Use for buildings, coasts, venues, interiors, product spaces, or game scenes; not talking-head ads, static showcases, ordinary drone footage, or montage trailers.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["first-person FPV","FPV tour","character-follow FPV","FPV flythrough","FPV venue tour","environment flythrough","flying character tour","paper airplane FPV","low-water skim","gap threading","route-driven FPV","aerial scene showcase"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"第一视角 FPV 穿越生成","tagEn":"Professional Film","tagCn":"专业影视","completeTagsEn":["Professional Film / Creative Generation"],"completeTagsCn":["专业影视 / 创作生成"],"descEn":"This Skill is for users who need a first-person FPV one-take video to showcase an uploaded scene, while optionally using an uploaded character as a visible motion guide. The user can provide a character image, scene image, architecture/interior/landscape reference, or an optional motion reference video. The Skill then analyzes spatial depth, water surfaces, traversable gaps, obstacles, and route flow, chooses route geometry based on the scene, only adds low-water skimming when the scene truly supports it, and plans a one-take route with close passes, banking, dives, gap threading, and strong parallax. It is suitable for scene showcases, tourism/architecture/space promotion, and character-guided flight, but not for talking-head videos, static product ads, generic drone footage, or cut-heavy montage trailers.","descCn":"此 Skill 面向需要用第一人称 FPV 一镜到底展示上传场景的用户，同时可让上传角色作为持续可见的飞行动线引导物。用户可以提供角色图、场景图、建筑/室内/景观参考，或额外运镜参考视频；随后分析空间深度、水面、可穿越缝隙、障碍和动线，根据场景决定路线几何，只在场景真的支持时安排贴水掠飞，再规划近距离擦过、侧倾、俯冲、穿缝和强视差的一镜到底路线。适合场景展示、文旅/建筑/空间宣传和角色引导式飞行，不适用于真人口播、静态产品广告、通用航拍素材或频繁切镜的蒙太奇预告片。","tools":[],"skillType":"skill","sortWeight":480,"downloads":2888,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-14-21/video/1786713892104144460-1786713892097.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"pov-short-film-generator","version":"1.0.0","hash":"sha256:e21f0b7dcaa88f8a04625778ae0c50f9c0c996ef5ada362bf18ae965e3e10405","summary":"Plan strict first-person POV immersive short films.","summaryZh":"将创意、剧本或参考素材规划为严格主角眼睛视角的沉浸式短片制作方案。","description":"Validate and produce strict immersive first-person POV short-film workflows from concepts, scripts, reference images, or existing clips. Use for protagonist-eye short dramas, subjective-camera suspense, survival, romance, horror, workplace pressure, realistic daily-life POV, storyboard planning, clip prompts, and QC. Boundary: every shot must be visible through the protagonist’s eyes; no protagonist face, third-person protagonist shot, product-only showcase, or long-series planning.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":["first-person POV","POV short film","subjective camera","immersive short film","POV drama","protagonist-eye view","first-person short","POV video","第一人称POV","POV短片","主观视角视频","沉浸式短片","第一视角短剧","POV短剧","眼睛视角","第一视角"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"第一视角短片生成","tagEn":"Short Drama","tagCn":"短剧","completeTagsEn":["Short Drama / Planning","Short Drama / Creative Generation"],"completeTagsCn":["短剧漫剧 / 计划制定","短剧漫剧 / 创作生成"],"descEn":"This Skill turns concepts, scripts, reference images, or existing clips into strict first-person POV immersive short-film workflows. It locks the protagonist-eye viewpoint, prevents third-person protagonist shots and face exposure, physicalizes visual style choices, and guides specification writing, POV asset planning, storyboard design, clip prompts, quality checks, and assembly guidance for compact short-form scenes.","descCn":"面向制作悬疑、生存、恋爱、恐怖、职场或现实题材第一视角短片的创作者。用户提供创意、剧本、参考图或既有片段后，Skill 会锁定主角眼睛视角，把抽象风格转化为可执行的画面规则，依次完成规格书、第一视角资产规划、分镜设计、镜头提示词、质量检查与组装建议。最终交付可执行的沉浸式短片制作方案，并持续排除主角正脸、第三人称主角镜头和纯产品展示；适合紧凑短片，不负责长剧集统筹。","tools":[],"skillType":"skill","sortWeight":490,"downloads":2728,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620882582516275-1786620882576.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"suspense-title-sequence-generator","version":"1.0.0","hash":"sha256:9be4112da21b0b5b8e511d2cc5a47492600baf8fc18a804f7ff8b3b97b862d96","summary":"Anchors keyframes and credits to generate a suspense-aesthetic film opening.","summaryZh":"根据电影设定与参考素材锚定关键帧和英文署名，生成十五秒复古悬疑美学片头成片。","description":"Create a 15-second suspense-aesthetic title sequence from a film type, title, logline, cast, optional credits, and reference images. Confirm reference roles, design six keyframes, lock readable English credits and beat-driven transitions, then generate a 16:9 retro suspense-jazz opening. Preserve pop-art collage, hard-edge silhouettes, limited stylized face reveals, and dense source-prompt continuity. Use for film openings, not full trailers, long narratives, or celebrity replication.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["suspense aesthetic title sequence","film title sequence","suspense crime intro","jazz noir trailer","电影片头","悬疑美学片头","悬疑犯罪片头","复古动画片头"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"悬疑电影片头生成","tagEn":"Video","tagCn":"视频","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"For creators making suspense, crime, noir, or spy-style title sequences, this Skill takes film type, title, logline, cast, optional credits, and reference images. It confirms reference roles, anchors six keyframes, designs readable English credits and beat-driven motion, then generates a 15-second 16:9 suspense-aesthetic opening. It supports retro pop-art collage, hard-edge silhouettes, and limited stylized face reveals, not full trailers or IP/celebrity replication.","descCn":"面向制作悬疑、犯罪、黑色电影或间谍风片头的创作者。用户提供电影类型、片名、一句话剧情、演员表、可选署名和参考图后，Skill 会先确认素材角色，再锚定六张关键帧，设计清晰可读的英文署名与节拍动效，最后生成十五秒、十六比九的复古悬疑美学片头。成片强调波普拼贴、硬边剪影、有限人物露出和爵士悬疑节奏，适合作为电影或短片开场，不用于完整剧情预告、长篇叙事或真实 IP 与名人复刻。","tools":[],"skillType":"skill","sortWeight":500,"downloads":2625,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620885996021233-1786620885990.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"micro-expression-video-generator","version":"1.0.1","hash":"sha256:81d9aaffc3134e6e886706d92d4e126e21d6fbc7c94962e5c7a59bb1b3d4774e","summary":"Turns a character image, script, or emotion phrase into timed micro-expression prompts for believable, restrained screen performances.","summaryZh":"根据角色图、剧本或情绪短语设计分时微表情提示词，用于塑造自然可信的角色表演。","description":"Turn a character image, script excerpt, or emotion cue into timed micro-expression direction and an optional clip. Use for nuanced screen acting; not scripts, rigging, lip sync, or compositing.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["micro-expression","character acting prompt","nuanced emotion performance","character performance","acting prompt","emotional acting","tearful composure"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"人物微表情视频生成","tagEn":"Professional Film","tagCn":"专业影视","completeTagsEn":["Professional Film / Planning"],"completeTagsCn":["专业影视 / 计划制定"],"descEn":"This Skill helps filmmakers and character-video creators turn a character image, script segment, or short emotion phrase into a production-ready micro-expression prompt pack. It preserves the approved identity, scene, dialogue, blocking, and camera intent, then asks for duration and intensity before selecting emotion routes, visible facial and body cues, timing density, and continuity guardrails. The user reviews the prompt before video generation; MiniMax-H3 is the default, while another explicitly requested model is used only after a capability check. The final output is a concise, time-aware acting prompt and, after approval, an optional generated performance clip. It is best for restrained, tearful, hesitant, angry, affectionate, or quietly decisive moments, not full scriptwriting, facial rigging, lip-sync editing, diagnosis, or final compositing.","descCn":"本 Skill 面向影视创作者和角色视频制作者，将角色图片、剧本片段或一句情绪短语转化为可执行的微表情表演提示词。流程先保护已确认的角色身份、场景、台词、调度和运镜，再确认镜头时长与表演强度，按需组合情绪路线、可见的面部与身体细节、节奏密度和镜头连续性约束；用户审阅提示词后，默认使用 MiniMax-H3 生成表演片段，明确指定其他模型时先做能力检查。最终交付分时表演提示词及可选视频片段，适合克制、含泪强撑、犹豫、愤怒、心动和冷静决绝等复杂表演，不替代完整编剧、医学判断、面部绑定、口型编辑或最终合成。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":510,"downloads":2617,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-14-21/video/1786713874745782855-1786713874739.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"1.0.1"},{"name":"short-drama-multilingual-dubbing","version":"1.0.0","hash":"sha256:2ce8ac9990dd12e41e5428b671fb740b941dddf4c756028ee56325dc3f3462c7","summary":"Re-localize video dialogue into natural dubbing in any target language.","summaryZh":"分离人声与背景音并逐句复刻音色和节奏，输出目标语言配音视频与字幕文件。","description":"Localize uploaded short dramas into another language. Prefer a clean voice track and subtitle file; otherwise extract audio, separate voice and background, transcribe dialogue, and confirm the target. Build a dubbing table, approve the wording, then replicate source-line timbre, emotion, rhythm, pauses, and duration, using official voices as fallback. Rebuild subtitles from the table and mux the new mix onto a silent copy. Use for dialogue localization, not story rewriting or visual redesign.","tags":["Sound & Music","Creative & Experimental"],"tagsCn":["音频音乐","创意实验"],"creator":"","triggerWords":["short drama multilingual dubbing","localized dubbing","video dubbing localization","dialogue localization sheet","voice/background separation","audio replication dubbing","subtitle file dubbing","短剧多语言出海配音","本土化配音","多语言配音","人声分离","音频复刻","对话翻译配音"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"短剧多语言出海配音","tagEn":"Sound & Music","tagCn":"音频音乐","completeTagsEn":["Sound & Music / Post-production","Creative & Experimental / Post-production"],"completeTagsCn":["音频音乐 / 后期处理","创意实验 / 后期处理"],"descEn":"Re-localizes uploaded videos into dubbed versions in the user’s chosen target language. It extracts audio, separates voice and background stems, recognizes dialogue, and creates a dubbing table. It recommends clean voice tracks and subtitle files as sources for dialogue text and timing; otherwise it confirms extraction and target language first. After approval, it prioritizes line-level audio replication to preserve timbre, emotion, rhythm, pauses, intensity, and duration, then outputs the localized dubbed video and subtitle file.","descCn":"将用户上传视频重制为目标语言的本土化配音视频：先提取音轨并分离人声与背景声，识别对白并生成配音表；建议提供干净人声轨和字幕文件作为文字与时间来源，若无则确认是否从视频提取并选择目标语言。确认后优先逐句音频复刻，保留音色、情绪、节奏、停顿、强度和时长，最终输出本土化配音视频与字幕文件。适合短剧与剧情对白的多语言本地化，不负责改写原剧情或重新设计画面内容。","tools":[],"skillType":"skill","sortWeight":520,"downloads":2381,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620884229238303-1786620884220.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"lip-product-ad-generator","version":"1.0.0","hash":"sha256:1afe4f3b4b2a39ccfe1b40bb85562dcb310673b68c192d57d04df9ebceb53675","summary":"Builds a 30s horizontal lip-product TVC in a fixed order, requiring model portrait, 3×3 still-life, and storyboard files before video generation, while using the still-life only as storyboard reference and keeping interactions natural.","summaryZh":"按固定顺序生成横屏30秒唇部彩妆TVC，先完成模特图、3×3静物图和脚本文件再生成视频，并强化自然的人物与产品、道具互动。","description":"Create a horizontal lip-product TVC from references or from scratch. Produce the model portrait, 3×3 product still life, and storyboard in that order; after approval, generate H3 shot segments, background music, and the final composite. Preserve model identity, packaging, texture, readable copy, and natural interaction with products and props, with segment-level retries. Use for lipstick, gloss, tint, mud, oil, and glaze ads, not skincare, eye makeup, or generic beauty reels.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["lip product TVC","lipstick ad","lip gloss TVC","lip tint","lip mud","lip oil","lip beauty","lipstick commercial","horizontal 30s","background music"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"唇部彩妆广告生成","tagEn":"Creative & Experimental","tagCn":"创意实验","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"For creators making lipstick, lip gloss, lip tint, lip mud, lip oil, lip glaze, or other lip-product commercials from scratch or from uploaded references. It follows a fixed workflow: generate the model portrait, generate a 3×3 product still-life using the canvas still-life prompt, write or refine the storyboard, produce storyboard-accurate shot-level video segments only after all three files exist, using the still-life only as storyboard reference and requiring natural model/product/prop interaction beats, add background audio, and then composite and validate the final 30s TVC. It keeps model identity and product presentation consistent, supports segment-level regeneration, and stays brand-agnostic, texture-agnostic, packaging-agnostic, and lip-format-agnostic when the references are broad. It is best for H3-style horizontal beauty ads and does not cover generic beauty reels, skincare, eye makeup, or silent video-only delivery.","descCn":"面向需要从零开始制作口红、唇釉、唇泥、唇蜜、唇油、唇釉唇泥综合类或其他唇部广告的创作者。它固定按照模特图、3×3 静物图、分镜脚本、脚本级小分段视频、背景音乐、最终合成校验的顺序执行；静物图必须严格跟随画布静物图提示词，视频必须在模特图、3×3静物图和脚本文件完成后生成，静物图仅作分镜参考，视频按脚本拆段，并强制加入自然灵动的人物和产品互动、产品质感与道具对比、人物和道具互动镜头，保证模特身份和产品呈现一致，支持按段重做，也能在参考较泛时保持品牌、质地、包装和产品形式都可泛化。它最适合 H3 横屏美妆广告，不适合泛美妆混剪、护肤、眼妆或无背景音的纯视频交付。","tools":[],"skillType":"skill","sortWeight":530,"downloads":2307,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620871400679467-1786620871394.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"silkscreen-decorative-illustration-short","version":"1.0.1","hash":"sha256:c529836ea19cb003bf55750b32869694daafeed79fbb0adf760c3aa054898290","summary":"Turns a refined silkscreen-decorative illustration theme into a restrained TVC workflow that follows the user's language and adds clean SFX assembly plus an optional same-duration instrumental BGM version.","summaryZh":"将丝印装饰插画主题转成同步音效短片，并可追加同等时长的器乐BGM版，适合奢华叙事广告。","description":"Create luxury scarf-illustration style short videos with refined TVC pacing: fine Ligne Claire linework, dense decorative narrative compositions, ivory or lime-white grounds, balanced warm and cool jewel accents, fixed wide establishing shots plus close follow-up shots, and synchronized natural sound effects only. Use for luxury scarf-illustration mood, decorative narrative TVC, Chinese courtyard tea, palace banquet, Paris apartment daily scene, animal-and-character luxury illustration video. Treat premium silkscreen-decorative references as non-branded visual shorthand only. Not for copying protected brand logos, existing scarf designs, dialogue-driven drama, BGM music videos, or ordinary realistic ads.","tags":["Creative & Experimental","Commercial Ad"],"tagsCn":["创意实验","商业广告"],"creator":"","triggerWords":["silkscreen-decorative-illustration-short","silkscreen decorative short","luxury scarf illustration","scarf-style TVC","Ligne Claire","decorative narrative short"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"丝印装饰插画风格短片","tagEn":"Creative & Experimental","tagCn":"创意实验","completeTagsEn":["Creative & Experimental / Creative Generation","Commercial Ad / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成","商业广告 / 创作生成"],"descEn":"Creates a refined silkscreen-decorative illustration short with restrained TVC pacing. It confirms the final video spec, analyzes references, designs key elements and paired wide/close shots, generates images and videos, adds synchronized natural sound effects, assembles a clean SFX cut, and optionally mixes a same-duration instrumental BGM version. MiniMax-H3 is the default; a user-selected other model is checked before use.","descCn":"本 Skill 面向想制作奢侈品创意广告片质感插画短视频的创作者，强调精细均匀线稿、高密度装饰叙事、冷暖宝石色和平面丝印质感。流程会先确认 Final Video Spec，分析参考图，设计 key element 与固定的全景建立镜头、局部近景跟随镜头，生成元素图和逐镜视频，添加与动作同步的自然音效并完成硬切组接；用户需要时，再生成同等时长的器乐 BGM 版。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败后按原因重试并切换可用模型。","tools":["question","hub_generate_image","hub_generate_video","hub_generate_music","hub_video_edit","hub_analyse_media","hub_save_file_to_session"],"skillType":"skill","sortWeight":540,"downloads":2296,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/8f3ee671-c950-440a-8ea2-e23fc61a3af8.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"vox-style-video-generator","version":"1.0.1","hash":"sha256:e9d1e0acd27d982164eaa740fb86924126fd4288db3bde9963fd06a000f3657f","summary":"Turn a topic or report into an approved Vox-style collage explainer with narration, subtitles, and assembled motion.","summaryZh":"根据主题或资料先做方案确认，再生成带旁白、字幕与拼贴动效的科普解释视频。","description":"Create Vox-style explainers from a topic, article, or report. Confirm the plan, then generate collage keyframes, audio, subtitles, and a reviewed final video for science or documentary shorts.","tags":["Education","Creative & Experimental"],"tagsCn":["教育","创意实验"],"creator":"","triggerWords":["explainer video","Vox-style video","science explainer","documentary short","educational video essay","automated video essay","one prompt to video","keyframe-to-animation"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"Vox风格科普动画","tagEn":"Education","tagCn":"教育","completeTagsEn":["Education / Creative Generation","Creative & Experimental / Creative Generation"],"completeTagsCn":["教育 / 创作生成","创意实验 / 创作生成"],"descEn":"For creators turning a topic, article, or research report into a clear explainer video. Provide source text or reference media; the Skill confirms language, audience, duration, aspect ratio, and subtitle needs, then prepares a production plan, script, storyboard, shot details, and full voiceover for approval. After approval it creates a visual anchor, consistent paper-collage keyframes, controlled animation, narration, optional subtitles, BGM, and a reviewed final assembly. The workflow preserves approval gates before costly generation and keeps audio aligned to each visual beat. Best for science explainers, documentary shorts, and educational video essays; not for fictional long-form stories, pure character animation, or unapproved batch generation.","descCn":"面向需要把知识主题、文章或研究报告转成解释型视频的创作者。用户提供主题、正文、资料或参考素材后，Skill 先确认语言、时长、画幅与受众，生成制作方案、脚本、分镜、镜头细节和旁白稿供确认，再制作主视觉、关键帧与纸艺拼贴动画，最后匹配统一口播、字幕和背景音乐并完成合成质检。适合科普、纪录短片和教育视频论文；不适合虚构长剧情、纯角色动画或未经确认的连续批量生成。","tools":[],"skillType":"skill","sortWeight":550,"downloads":2350,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/3b455fbe-7608-4804-80c7-34411049d7e2.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"co-op-game-intro-generator","version":"0.2.1","hash":"sha256:a3b4c502e7251d687944260daf4afacce9d9a5cbb71c9db2286cf8dce6c36f04","summary":"Create a reusable co-op game menu intro from player names, visual style, character references, and one approved confirmation image.","summaryZh":"根据玩家名称、视觉风格和角色参考图，经首图确认后生成双人游戏菜单开场视频。","description":"Create a two-player co-op game menu or opening animation from two player names, a title, a visual style, and optional character references. Lock identity cues, generate one menu approval image, then use it to rebuild character, interface copy, timing, and motion instructions before producing the final H3 video. Use for game concepts, character-led menus, and social intros, not playable game development, exact logo replication, complex multi-page interfaces, or character-free title sequences.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["co-op intro","game menu","two-player game intro","H3 game intro","two-player game opening","game main menu","player names","game title"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"双人游戏开场视频生成器","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"This Skill is for creators who want a repeatable, platform-ready workflow for two-player co-op game menu intros and opening animations. The user provides PLAYER 1 and PLAYER 2 names, a game title, a desired visual style, and optional character reference images. The workflow first maps identity cues without inheriting unwanted source-image realism, fills a fixed confirmation-image template, generates one 16:9 menu approval image, waits for user approval, then refills an H3 video prompt from the approved image, player data, UI copy, timing events, motion directions, and negative constraints. The final deliverable is a co-op game intro video featuring two characters, player cards, menu interaction, equipment configuration, loading, and world-entry motion. It is best for game concept videos, social media openings, and character-led menu animation, not for playable game development, complex multi-page UI systems, exact commercial logo replication, or generic character-free title sequences.","descCn":"本 Skill 面向需要反复制作双人合作游戏主菜单或开场动画的创作者，并已整理为平台可发布结构。用户提供 PLAYER 1 与 PLAYER 2 名称、游戏标题、目标视觉风格，并可上传角色参考图。流程会先提取身份锚点但避免继承原图写实质感，再填写固定确认首图模板，生成一张 16:9 菜单确认图；用户批准后，基于确认图、玩家信息、UI 文案、时间线事件、运动方向和负面约束回填 H3 视频提示词。最终输出包含双人角色、玩家信息卡、菜单交互、装备配置、加载和进入游戏世界的开场视频。适用于游戏概念展示、社交内容和角色化菜单动效，不处理可玩游戏开发、复杂多页面 UI、精确商业 Logo 复刻或无角色通用片头。","tools":["question","hub_generate_image","hub_generate_video","hub_analyse_media","hub_canvas_get_node","hub_save_file_to_session"],"skillType":"skill","sortWeight":560,"downloads":2274,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620867045038554-1786620867039.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"h3-prompt-expert","version":"1.0.2","hash":"sha256:20281d1012ef6d272f5562b0a84a25f82540eb6abbffadfbb56b40ea6e3a20c7","summary":"Evaluates video intent and references, turning scattered inputs into controllable multimodal video generation instructions.","summaryZh":"判断视频需求与参考素材，将零散输入整理成可控的多模态视频生成指令。","description":"Turn video ideas, assets, keyframes, references, or edit requests into official MiniMax H3 prompts. Use for H3 prompt writing or rewriting; not for other models, full scripts, or long-form planning.","tags":["Platform Tooling"],"tagsCn":["平台工具"],"creator":"","triggerWords":["H3 prompt","MiniMax H3 prompt","multimodal video prompt","reference image prompt","keyframe prompt","video edit prompt"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"H3 提示词专家","tagEn":"Platform Tooling","tagCn":"平台工具","completeTagsEn":["Platform Tooling"],"completeTagsCn":["平台工具"],"descEn":"Turns rough video ideas, asset lists, keyframes, multimodal references, and edit instructions into production-ready MiniMax-H3 prompts. It diagnoses the request, checks reference fit, asks focused choice-card questions, and outputs official H3 structures for reference roles, UI/text control, keyframes, music, dialogue, and edit preservation. It does not replace full scriptwriting, storyboards, long workflows, or prompts for other models.","descCn":"这是一个本地 Skill，用于把粗略视频创意、素材列表、关键帧需求、多模态参考和视频编辑指令转成可生产使用的 MiniMax H3 提示词。它会先识别任务类型，评估用户文字和参考图是否满足目标，缺少会改变 prompt 骨架的信息时用选项提问，并在用户选择后只输出官方 H3 格式，如 integrated_multimodal_description 或 subject_definitions 与 retention_analysis。它适合 case 级 H3 提示词、参考素材角色映射、UI/文字/布局控制、关键帧对齐、音乐卡点/对白和编辑保留；不替代完整剧本、分镜规划、长期项目流程或其他视频模型提示词。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":570,"downloads":2362,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/adfe03a1-cff0-4380-a45c-121e39621ae1.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"1.0.2"},{"name":"handdrawn-live-video-generator","version":"1.0.8","hash":"sha256:6adba155e309b90dd97267ace4a9c51fece0e328cb98f25e7ad31865d010f8fd","summary":"Create H3-ready prompts for hand-drawn animation interacting with live-action scenes.","summaryZh":"基于场景创意，生成手绘动画与实拍接触变形的15秒视频，适用于创意短片制作。","description":"Create a surreal 15-second 16:9 short blending rough glowing hand-drawn animation with live action. Clarify the contact object or hand, mood, language, and style; design continuous morphing, an escape route, and delayed handheld pursuit; then write the prompt in the user's language and generate with H3 after approval. Check contact realism, camera delay, rough stroke texture, and a non-horror tone. Use for single scenes, not polished CG, jump scares, plush characters, or multi-scene cuts.","tags":["Creative & Experimental","Animation"],"tagsCn":["创意实验","动画"],"creator":"","triggerWords":["hand-drawn glowing animation fusion","15-second morphing chase video prompt","H3 video generation","video prompt","hand-drawn animation touching real objects","crayon-chalk texture","multilingual video prompts"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"手绘实拍融合视频生成器","tagEn":"Creative & Experimental","tagCn":"创意实验","completeTagsEn":["Creative & Experimental / Creative Generation","Animation / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成","动画 / 创作生成"],"descEn":"For creators making surreal short videos that blend rough glowing hand-drawn animation with real-world footage. Users provide a scene idea, contact object or hand, desired mood, and optional language or style constraints. The Skill clarifies the interaction, designs continuous morphing and chase action, writes a reusable 15-second 16:9 video prompt in the user's language, recommends MiniMax H3 after confirmation, and checks contact realism, camera delay, texture, and non-horror tone. Best for single-scene creative clips, not polished CG, horror jump scares, plush characters, or multi-scene editing.","descCn":"面向希望制作手绘动画与实拍空间融合短片的创作者。用户提供场景创意、接触对象或手部动作、情绪氛围和可选风格限制。Skill 会明确触碰关系，设计连续变形、逃跑路线和慢半拍手持追拍，整理为用户输入语言的15秒16:9视频提示词；确认后建议使用 MiniMax H3 生成，并检查接触真实感、镜头延迟、笔触质感和非恐怖调性。适合单场景创意视频，不适合精致CG、恐怖跳吓、毛绒角色或多场景剪辑。","tools":["question","hub_generate_video","hub_analyse_media","hub_save_file_to_session"],"skillType":"skill","badges":["silent-install"],"sortWeight":580,"downloads":2191,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-13-19/video/1786620873229141646-1786620873219.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":true,"updateAvailable":false,"installedVersion":"1.0.8"},{"name":"papercraft-stop-motion-explainer","version":"0.6.9","hash":"sha256:915546a267fc6382df3d274a696b11c988dc83d4d3449f13456ad71c5e5a9e4a","summary":"Build a production-ready papercraft stop-motion explainer package from a knowledge topic.","summaryZh":"根据知识主题，完成纸偶、分层布景、分镜与提示词设计，输出可执行的纸艺定格科普视频创作包。","description":"For creators explaining science, education, or general knowledge through tactile handmade papercraft visuals. Users provide a topic, core knowledge points, or source material and may specify audience, duration, aspect ratio, and deliverable type. The Skill extracts the learning goal and visual metaphor, proposes creative directions, designs paper characters, layered diorama sets, and props, creates preview concepts plus image and video prompts, and plans storyboards, camera movement, transitions, and sound with staged approvals and review checklists. It outputs a production-ready papercraft stop-motion explainer package, or selected assets such as still prompts, image-series prompts, short-video prompts, or storyboards. Best for cut-paper, pop-up-book, layered diorama, and miniature stop-motion explainers; not for standard 2D cartoons, line doodles, live action, or explainers without a paper-art look.","tags":["Education","Animation","Creative & Experimental"],"tagsCn":["教育","动画","创意实验"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"纸艺定格科普视频生成器","tagEn":"Education","tagCn":"教育","completeTagsEn":["Education / Creative Generation","Animation / Creative Generation","Creative & Experimental / Creative Generation"],"completeTagsCn":["教育 / 创作生成","动画 / 创作生成","创意实验 / 创作生成"],"descEn":"For creators explaining science, education, or general knowledge through tactile handmade papercraft visuals. Users provide a topic, core knowledge points, or source material and may specify audience, duration, aspect ratio, and deliverable type. The Skill extracts the learning goal and visual metaphor, proposes creative directions, designs paper characters, layered diorama sets, and props, creates preview concepts plus image and video prompts, and plans storyboards, camera movement, transitions, and sound with staged approvals and review checklists. It outputs a production-ready papercraft stop-motion explainer package, or selected assets such as still prompts, image-series prompts, short-video prompts, or storyboards. Best for cut-paper, pop-up-book, layered diorama, and miniature stop-motion explainers; not for standard 2D cartoons, line doodles, live action, or explainers without a paper-art look.","descCn":"面向希望用手工纸艺视觉讲解科学、教育或泛知识内容的创作者。用户需提供主题、核心知识点或原始材料，并可指定受众、时长、画幅和交付类型。Skill 会先提炼学习目标与视觉隐喻，提出创意方向，设计纸偶角色、分层纸雕布景和道具资产，制作预览概念、图像与视频提示词，规划分镜、运镜、转场和声音，并通过阶段确认与审核清单控制质量。最终输出可直接进入制作的纸艺定格科普视频创作包，也可按需交付单图、系列图、短视频提示词或分镜。适用于纸雕、剪纸、立体书和微缩定格讲解，不适用于普通二维卡通、线稿涂鸦、真人写实视频或无纸艺质感的标准科普。","tools":[],"skillType":"skill","sortWeight":590,"downloads":2057,"coverUrl":"https://cdn.hailuoai.video/open-hailuo-video-web/public_assets/53500488-f43a-4931-a171-5cb2f189b4d5.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"transcript-broll-planner","version":"1.0.1","hash":"sha256:1528a373e57f159657a19cc3308a1f80b1890b37ccc0fb9fc5998990c2f0180a","summary":"Turns transcript-led narration into reviewable B-roll plans and approval-gated video shots for knowledge and talking-head edits.","summaryZh":"把口播逐字稿按语义拆成可审片的 B-roll 镜头方案，并在用户确认后生成已批准的视频素材。","description":"Plan transcript-driven B-roll for talking-head and knowledge videos. Use when the user provides a script, voiceover, data table, screenshots, or references and needs semantic segmentation, shot selection, missing-material checks, bilingual on-screen text control, continuous one-take design, a reviewable B-roll plan, and approval-gated generation. Trigger for transcript B-roll planning, talking-head B-roll, voiceover B-roll, narration B-roll, or B-roll shot planning.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["transcript B-roll planner","B-roll planning","transcript-driven B-roll","talking-head B-roll","voiceover B-roll","narration B-roll","B-roll shot planning"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"口播视频B-roll规划器","tagEn":"Professional Film","tagCn":"专业影视","completeTagsEn":["Professional Film / Full Production"],"completeTagsCn":["专业影视 / 成片制作"],"descEn":"For creators turning a transcript, voiceover, data-led narration, or talking-head script into reviewable B-roll and approved video assets. Provide the full script and optional logos, screenshots, documents, charts, product images, data, or visual references. The workflow segments meaning, decides whether each unit keeps A-roll, reuses supplied evidence, uses motion typography, becomes visual B-roll, or joins a continuous take, then selects a template and flags missing material. It delivers a shot plan with screen-language, continuity, and deterministic-production notes before generation, then produces only approved shots. Best for knowledge videos, explainers, market analysis, and transcript-led edits; not subtitle-only export, simple trimming, one-off image prompts, or unrelated long-form editing.","descCn":"面向把逐字稿、配音稿、数据型讲稿或口播脚本转成可审片 B-roll 和视频素材的创作者。用户提供完整脚本，可补充 Logo、截图、文档、图表、产品图、数据表或参考。流程先按语义拆分，判断各段应保留 A-roll、复用来源素材、使用动态文字、转为纯视觉副镜头或纳入连续长镜头，再选择模板并标出缺失素材。生成前交付包含屏幕文字语言、连续性和确定性制作说明的镜头计划，确认后仅生成已批准镜头。适合知识视频、讲解、市场分析和逐字稿驱动剪辑；不用于纯字幕导出、简单剪切、单次图片提示词或无关长篇剪辑。","tools":[],"skillType":"skill","sortWeight":600,"downloads":2051,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-14-21/video/1786713893768400104-1786713893760.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"brand-ad","version":"1.1.3","hash":"sha256:312e2a7ffb423dcc806bacbe13cf8114cd12146d25aed5b2e1a923f1765ca6dd","summary":"Produce a 15, 30, or 45-second official brand or product hero film","summaryZh":"把品牌事实与产品卖点转化为十五、三十或四十五秒轻量品牌广告成片","description":"Create one official brand or product-hero ad up to 45 seconds from verified facts, assets, audience, channel, copy, and sound needs. Not for creator reviews, unboxing, testimonials, or KOC content.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["brand ad","product ad","product hero film","official commercial","brand showcase","product commercial","brand process film"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"品牌广告","tagEn":"Commercial Ad / Full Production","tagCn":"商业广告 / 成片制作","completeTagsEn":["Commercial Ad / Full Production"],"completeTagsCn":["商业广告 / 成片制作"],"descEn":"For brands and product teams that need one lightweight official ad planned at 15, 30, or 45 seconds. Verifies brand facts, selects an internal creative route, compiles LLM-developed copy and art typography, builds timecoded prompts, generates independent video clips with native narration and effects, and concatenates the complete clips on canvas without forced duration edits.","descCn":"面向需要制作单条轻量品牌或产品广告的品牌与产品团队，将用户提供或经官方核验的品牌事实、产品资产、卖点和授权信息整理为广告方案。流程覆盖品牌与素材真实性核验、广告路线选择、视觉方向收敛、时间码分镜、屏幕文案与艺术字体编排、产品锚点和声音策略确认、独立片段生成、原生旁白与动作音效处理，再按实际顺序完整拼接并交付。支持十五、三十和四十五秒目标；身份、文字、包装、版权或最终时长有风险时暂停并请求补充，不把未经确认的宣传承诺当成事实。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":610,"downloads":1846,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"1.1.3"},{"name":"koc-video","version":"8.7.2","hash":"sha256:2d7effb70ff17a7d7ebe02cbc6a29a0d279b56697627ab137223e77746cee6d9","summary":"Generate creator-native social videos including product seeding, talking heads, testimonials, POV, and virtual hosts","summaryZh":"生成达人原生风格的社交种草视频，覆盖口播、测评、证言、无脸视角与数字人栏目","description":"Create creator-native social video from verified product facts, audience goals, and optional references. Delivers reviews, testimonials, POV, or virtual-host content; not official brand films.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["KOC video","UGC video","product seeding","creator review","testimonial video","faceless POV","hands-only demo","virtual host video"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"达人种草视频","tagEn":"Commercial Ad / Full Production","tagCn":"商业广告 / 成片制作","completeTagsEn":["Commercial Ad / Full Production"],"completeTagsCn":["商业广告 / 成片制作"],"descEn":"For brands and creators making creator-native social content: product or service seeding, creator talking heads, expert shares, customer testimonials, faceless POV, hands-only demos, digital humans, and account series formats. Built for Douyin, Xiaohongshu, TikTok, and Reels content that drives immersion, engagement, follows, inquiries, or conversion. Neither a physical product nor a real person on camera is required. Not for brand films, TVC, serialized drama, news explainers, or videos without an audience action goal.","descCn":"面向品牌与创作者的达人原生内容生产入口，生成产品或服务种草、创作者口播、专家分享、客户证言、无脸视角、手部实测、数字人与虚拟角色以及账号栏目等形态的社交视频。适用于抖音、小红书等平台中服务于代入、互动、关注、咨询或转化目标的内容，产品与真人出镜都不是前提条件。流程覆盖表达视角确认、创意路线选择、脚本与画面规划、视频生成与成片交付，不适用于品牌大片、连续剧情或没有受众行动目标的普通视频。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":620,"downloads":1945,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":true,"installedVersion":"8.7.1"},{"name":"line-doodle-explainer-generator","version":"1.0.3","hash":"sha256:52f252bb89fba59d84df8fe5b2271dc61229a16d4c15e0c82a8147ec519241aa","summary":"Generate line-doodle infographic explainer video plans and production assets from knowledge topics or lesson materials.","summaryZh":"根据知识主题或教材内容，生成线条小人图解科普视频方案与成片素材，适用于知识讲解短视频。","description":"Turn a topic or lesson into a minimalist line-doodle explainer plan and video, with narration, diagrams, storyboards, media production, assembly, and gated review. Not for exercises or cartoons.","tags":["Education"],"tagsCn":["教育"],"creator":"","triggerWords":["line doodle explainer","stick figure education video","minimalist line animation","infographic teaching video","science explainer video","knowledge explainer video","production plan"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"线条图解科普视频生成","tagEn":"Education","tagCn":"教育","completeTagsEn":["Education / Intent Brainstorming","Education / Planning","Education / Creative Generation","Education / Post-production"],"completeTagsCn":["教育 / 意图脑爆","教育 / 计划制定","教育 / 创作生成","教育 / 后期处理"],"descEn":"Turns science, humanities, history, or lesson material into a line-doodle infographic explainer. The workflow clarifies the brief, extracts the core idea, writes narration, designs diagrams and a storyboard, plans assets, generates media, assembles the video, and reviews quality at approval gates. Best for educational knowledge shorts; not for live-action realism, complex narrative films, or classroom exercises.","descCn":"面向需要将知识内容制作成线条小人图解科普视频的用户，适用于科学、人文、历史、社会科学、教材片段或课程材料的视频化表达。Skill 会先梳理需求与视频方案，再提炼核心观点、撰写口播脚本、设计图解系统和分镜，随后规划素材、生成媒体、完成合成与质检，并在主要节点用选项卡确认。适合知识讲解和教育短视频，不适合真人写实、复杂剧情或课堂练习题生成，强调信息清晰与节奏易懂。","tools":[],"skillType":"skill","sortWeight":630,"downloads":1834,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/2990c6db-3f34-468d-a03c-47dbb9efd89d.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"ecommerce-design","version":"1.0.11","hash":"sha256:e1e579ec31fa8abb8e8c5c2b34fffd6b05404fbdc3356eaa13cc0e20046783cf","summary":"E-commerce design across 14 categories","summaryZh":"覆盖14品类的电商图设计","description":"Professional e-commerce image design skill. Covers 14 product categories (women's clothing / home decor / daily essentials / pets / baby & maternity / digital & 3C / jewelry & accessories / beauty & skincare / health supplements / personal care / men's clothing / footwear / bags & luggage / sports & outdoor) + 7 anti-failure iron laws + 3 workflow meta constraints + 6 standard image templates (white background / hero / detail / lifestyle / size reference / multi-angle) + Attribute Card / Identity Lock / Scene Lock consistency system + C1–C11 prompt clauses + R1–R3 regeneration clauses + 6-dimension weighted scoring system + 14 category weight overrides + compliance red lines (baby & maternity / health supplements / beauty / personal care / 3C / food / beverages) + platform specifications (Amazon / Taobao / JD.com / Xiaohongshu / Shopify etc.). Core capabilities: product-first, identity lock, cross-image consistency, compliance and credibility. Trigger: \"ecommerce\", \"product image\", \"product photo\", \"white background shot\", \"hero image\", \"detail page\", \"hero shot\", \"lifestyle\", \"detail shot\", \"macro\", \"close-up\", \"multi-angle\", \"size reference\", \"on-model shot\", \"on-foot shot\", \"in-hand shot\", \"apparel photo\", \"footwear photo\", \"bag photo\", \"electronics photo\", \"beauty photo\", \"supplement photo\", \"baby product photo\", \"pet product photo\", \"home decor photo\", \"packaging shot\", \"on-pack\", \"Amazon hero image\", \"product listing photo\", \"ecommerce image set\", \"make a product photo\", \"create hero images\", \"ecommerce photo suite\". NOT for: marketing posters (use poster-design) / anime merchandise (use anime-design) / film storyboards (use film-shot)","tags":["E-Commerce"],"tagsCn":["电商"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"电商图设计","tagEn":"E-Commerce / Image Generation","tagCn":"电商 / 图像生成","completeTagsEn":["E-Commerce / Creative Generation"],"completeTagsCn":["电商 / 创作生成"],"descEn":"For full e-commerce visual design across 14 verticals such as apparel, beauty, 3C electronics, baby, pets and home furnishing. Takes product info or a brief plus optional reference images as input. Produces white-background, hero, detail page, lifestyle, size-reference and multi-angle images compliant with Amazon, Taobao, JD, Shopify and Xiaohongshu platform specifications and guidelines.","descCn":"适用于电商商品图的全套视觉设计,覆盖女装、美妆、3C 数码、母婴、宠物、家居等 14 个垂直品类。输入商品信息或设计简报、可选参考图,输出白底图、hero 主图、详情页图、lifestyle 场景图、尺寸参考图、多角度展示图等完整商品图套件。全部适配 Amazon、淘宝、京东、Shopify、小红书等平台的规格与合规要求。","tools":[],"skillType":"skill","sortWeight":640,"downloads":1743,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"video-transcript","version":"0.2.1","hash":"sha256:0cf933db16a50c6aa9df57792de4248f64dc3fb4c47e32597cc3b2e9b19a2c06","summary":"Turn uploaded video or audio into timestamped Markdown transcripts, subtitle sources, summaries, and quote banks.","summaryZh":"将音视频转为带时间戳逐字稿。","description":"Transcribe uploaded videos or audio files into readable timestamped Markdown transcripts, subtitle files, key-point summaries, and reusable quotes. Use for video-to-transcript, audio transcription, podcast transcription, interview extraction, meeting recordings, searchable archives, subtitle generation, or turning media into structured notes. Default workflow: quickly extract/normalize audio with ffmpeg, run the built-in ASR subtitle generator, then format the transcript; local Whisper is a fallback only when the built-in ASR fails or the user explicitly requests offline/local transcription.","tags":["Education"],"tagsCn":["教育"],"creator":"","triggerWords":["video-transcript","transcribe","transcript","video transcript","audio transcript","speech-to-text","subtitles","captions","转录","视频转录","逐字稿","字幕","视频转文字","音频转文字","提取字幕"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"视频转录","tagEn":"Audio Content / Post-Production","tagCn":"音频内容 / 后期制作","completeTagsEn":["Education / Post-production"],"completeTagsCn":["教育 / 后期处理"],"descEn":"Video Transcript converts uploaded video or audio into readable transcripts. It first extracts and normalizes audio with ffmpeg, uses the built-in ASR subtitle generator as the primary path, and treats local Whisper as an optional fallback for offline or failed-ASR cases. It can produce timestamped Markdown, subtitle sources, key-point summaries, and reusable quote banks for videos, podcasts, interviews, and meetings.","descCn":"视频转录把上传的视频或音频整理成可阅读的逐字稿。默认先用 ffmpeg 快速抽取并标准化音频，再调用内置 ASR 生成字幕源；只有在主 ASR 失败、结果无效或用户明确要求本地离线时，才把 Whisper 作为兜底方案。适合视频字幕、播客转录、访谈整理、会议记录、关键观点和金句提取。","tools":[],"skillType":"skill","sortWeight":650,"downloads":1711,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"digital-product-promo-generator","version":"1.0.0","hash":"sha256:c8d883de54c9848ac82283b80c1a3a1ba784bec46cf2fdcd731c31007a458039","summary":"Turns real product pages, frontend projects, screenshots, and supplied motion references into reviewable cinematic promos with accurate UI, purposeful camera movement, rhythmic editing, and sound design.","summaryZh":"把真实产品页面、前端项目与用户动效参考转成可审片的电影感宣传片，准确呈现界面、运镜、节奏和声音。","description":"Turn a product, webpage, frontend project, or shot reference into a cinematic promo. Collect the real URL, screenshots, brand assets, duration, and ratio; analyze its visual language; propose a reviewable shot, rhythm, and sound plan; then generate and edit the approved video. Treat Ink Press or shot-card references as style constraints, not bundled assets. Use for launch films, feature demos, and product storytelling, not generic editing or unrelated motion art.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["digital product promo","product video","promo video","shot card","frontend promo","webpage video","数字产品宣传片","网页产品视频"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"数字产品宣传片生成","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Full Production"],"completeTagsCn":["商业广告 / 成片制作"],"descEn":"This Skill helps product teams and creators turn a webpage, frontend project, screenshots, screen recordings, or a user-supplied motion reference into a cinematic product promo. It first collects the real product source, brand assets, target duration, aspect ratio, audience, and delivery constraints, then analyzes the interface's layout, typography, colors, copy, and interaction logic. Before generation, it presents a reviewable narrative, shot, camera, transition, text, rhythm, and sound plan while identifying any missing source material. After approval, it generates the confirmed shots with MiniMax-H3 by default, edits them with the approved audio, and checks continuity, text accuracy, beat alignment, and clean opening and end frames. User-supplied Ink Press examples or shot cards can guide the result, but no bundled template, code project, or audio pack is assumed. It is intended for product launch films, feature demos, and web or desktop product storytelling, not generic editing, talking-head cleanup, or unrelated motion art.","descCn":"本 Skill 面向需要制作产品宣传片、发布短片或功能演示的产品团队与创作者，可将网页、前端项目、截图、录屏或用户提供的动效参考转化为电影感成片。流程先收集真实产品来源、品牌素材、目标时长、画幅、受众与交付约束，再分析界面的布局、字体、颜色、文案和交互逻辑；生成前提供可审阅的叙事、镜头、运镜、转场、屏幕文字、节奏和声音方案，并明确缺失素材。用户确认后默认使用 MiniMax-H3 生成已批准镜头，结合确认的音频完成剪辑，再检查连续性、文字准确性、节奏卡点和干净首尾帧。用户提供的 Ink Press 示例或镜头卡可作为风格参考，但不假设 Skill 内置模板、代码工程或音频包。适合 Web 与桌面产品叙事，不用于通用剪辑、口播清理或无关动效创作。","tools":["question","webfetch","hub_analyse_media","hub_generate_video","hub_generate_audio","hub_video_edit","hub_save_file_to_session"],"skillType":"skill","sortWeight":660,"downloads":1728,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-14-21/video/1786713897391782105-1786713897381.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"fashion-scene-video-planner","version":"1.0.3","hash":"sha256:a70d90f3ef5318da4ead31efec6c87a1eed160183d2a0128717c6e9c39ff5238","summary":"Turns garment references into staged showcase videos by locking specifications, confirming assets, previewing storyboards, and delivering a 30-second cut.","summaryZh":"把服装参考分阶段编排成时装展示视频，锁定规格、确认资产并输出30秒成片。","description":"Plan a repeatable fashion showcase from garment references or a clothing brief, including outfit assets, scenes, storyboards, video, music, and assembly. Use for lookbooks and apparel films.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["fashion showcase","lookbook","apparel workflow","clothing short","fashion video","showcase video"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"时装场景展示视频","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成","电商 / 创作生成"],"descEn":"For fashion teams, designers, and creators who need a repeatable showcase pipeline from garment photos, lookbook references, or a clothing brief. The workflow locks a final specification, analyzes garment structure and fabric behavior, confirms technical outfit sheets, model cards, scenes, and a six-panel storyboard, then generates one complete 30-second showcase per outfit with music and assembly. MiniMax-H3 is the default video model; an explicitly selected alternative is checked for capability, retried once on failure, and then replaced with another available model. Best for runway-style showcases and lookbooks, not simple retouching, generic product ads, or unrelated narrative shorts.","descCn":"面向服装团队、设计师和创作者，把服装照片、Lookbook 参考或穿搭 brief 转成可复用的时装展示视频流程。系统先锁定最终规格，分析服装结构与面料物理，分步确认刀版图、模特服饰卡、场景候选和六宫格分镜，再按每套服装生成完整 30 秒成片，并完成配乐与组装。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。适合秀场展示、生活方式 Lookbook 和服装场景视频，不负责简单修图、泛品类广告或无关剧情短片。","tools":[],"skillType":"skill","sortWeight":670,"downloads":1678,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/3683f35f-cda6-4a86-a3c5-64d06173ed69.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"education-studio","version":"0.1.47","hash":"sha256:cdd007bf813efac0d76edd66e905fda375d6d3ba2022768314002519ea921d09","summary":"Turn topics, questions, or course material into lesson designs, exercises, and complete teaching videos","summaryZh":"把主题、题目或课程素材转化为教学设计、练习题组与完整教学视频的一体化工作室","description":"Turn a topic, question, lesson recording, transcript, or course material into a lesson plan, exercises, handout, courseware, or educational video. Not for generic editing without a teaching goal.","tags":["Education"],"tagsCn":["教育"],"creator":"MiniMax","triggerWords":["educational video","teaching video","lesson plan","course material","quiz creation","science explainer","training content","course transcript"],"guidePrompt":"告诉我你想讲解的主题、题目或课程素材，我会通过可点击选项引导你完成教学设计、分镜、H3 视频生成和成片。","guidePromptEn":"Share a topic, question, or course material. I will guide you through lesson design, shot planning, H3 video generation, and final delivery using clickable choices.","displayNameZh":"教育内容工作室","tagEn":"Education / Full Production","tagCn":"教育 / 成片制作","completeTagsEn":["Education / Full Production","Education / Planning"],"completeTagsCn":["教育 / 成片制作","教育 / 计划制定"],"descEn":"For teachers, course creators, and knowledge publishers who need to turn a topic, exam question, or raw course material into teachable content. Supports learning-objective design, explanations with worked examples, classroom activities, exercise sets with answer keys, courseware and handout structures, plus full video production from brief, word-for-word narration, per-shot teaching contracts, and storyboards to video generation, sound, subtitles, and final render. Also processes recordings and transcripts into clean lesson notes and knowledge slices.","descCn":"面向教师、课程创作者与知识内容团队，把主题、试题、课堂录音或教材素材加工成可教学、可学习、可评量的内容。支持学习目标与路径设计、讲解与例题、课堂活动、题组与答案解析、课件与讲义结构，并覆盖教学视频从需求简报、逐字旁白、逐镜教学合同、分镜规划到视频生成、声音、字幕与最终渲染的完整生产流程。同时支持把课堂录音整理为逐字稿、讲义与知识点切片，适用于课程视频、动态讲题、科普讲解与培训内容制作。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":680,"downloads":1738,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":true,"installedVersion":"0.1.46"},{"name":"poster-motion-generator","version":"1.0.1","hash":"sha256:169fd2509dfa46b3e4c1e6edadfc1ca0b3accf419aa2b94796550cf2818e44e9","summary":"Turns uploaded posters into single-take MiniMax H3 motion posters with ratio mapping, readable text, and timestamped motion prompts.","summaryZh":"将上传海报转为一镜到底动态视频，保留文字清晰、主体关系与最终版式，适合社媒传播。","description":"Poster Motion Generator turns an uploaded poster into an H3 single-take motion poster. It analyzes aspect ratio, subject, poster type, layout, visual hierarchy, typography, and motion potential, then offers 8 motion directions with Direction 1 Progressive Activation and Direction 2 Paper-Roll Reveal prioritized. The uploaded poster is the final lock-frame reference, not a required opening frame. Built in: motion-effect terms, direction-aware SFX wording, and timestamp storyboard prompt format. Trigger words: poster animation, poster-to-video, dynamic poster, animated poster, motion poster, poster awakening.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["Poster Motion Generator","poster animation","poster-to-video","dynamic poster","animated poster","motion poster","poster awakening","poster video","MiniMax H3"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"海报动态生成器","tagEn":"Creative & Experimental","tagCn":"创意实验","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"Turns an uploaded poster into a polished single-take motion video. The workflow analyzes ratio, layers, typography, anchors, depth, and motion potential, then creates reusable directions, a 2K 8–15-second specification, timestamped prompts, and synchronized sound cues before generation. MiniMax-H3 is the default video model; another model is used only after explicit user selection and capability checking.","descCn":"面向已有海报图片、希望生成精致动态海报视频的创作者。Skill 会先分析比例、海报类型、视觉层级、文字版式、锚点、景深和动效潜力，再输出可复用动态方向；上传海报作为最终锁版参考，随后制定 2K、8–15 秒一镜到底规格，编写带时间码的动效提示词与同步音效，并在确认后生成成片。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力再执行，强调文字可读、版式稳定、主体连贯与画面节奏。","tools":["question","hub_analyse_media","hub_generate_video","hub_save_file_to_session"],"skillType":"skill","sortWeight":690,"downloads":1604,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/ae64bf13-9da5-413c-b116-35a2348f2bd8.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"dot-matrix-brand-wordmark-motion","version":"1.0.0","hash":"sha256:6734fab784bfba7617a0e49dd24c30ab8ed4303b745c1c1b783dbaa5fcab5c04","summary":"Creates dot matrix brand wordmark motion from a supplied logo, brand name, and slogan, producing a horizontal brand plate, a rewritten motion prompt, and a dot matrix brand wordmark motion video.","summaryZh":"根据用户提供的品牌 Logo、品牌名与标语生成点阵品牌字形动效，输出横版品牌图、改写后的动效提示词与点阵品牌字形动效视频。","description":"Turn a supplied logo, brand-name image, or slogan visual into dot-matrix wordmark motion. Create a horizontal brand plate, rewrite the fixed prompt from visible or user-provided text, pause for approval, then generate an 8-second 16:9 video. Load references/chinese-prompts.md and preserve exact text, slogan, character anchors, and a clean tail frame. Use for brand openers and identity reels, not trademark design, slogan invention, or unsupported claims.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["logo slogan video","brand logo motion","logo to H3 video","brand logo animation","slogan brand video"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"点阵品牌字形动效","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成"],"descEn":"This Skill is designed for users who already have a logo, brand-name image, or slogan-containing brand visual and want a repeatable brand motion workflow rather than a one-off prompt. The user provides a logo image and may optionally describe the brand, but all final readable text must come from the image itself or explicit user-provided brand/slogan text. The workflow first creates a 16:9 horizontal brand-name plate with GPT Image, preserving the source image’s theme color and design background. It then reads that generated plate with the current model, treats the plate text as authoritative, extracts only the visible brand name and slogan, and rewrites a fixed MiniMax H motion prompt so the first, last, and prefix character logic matches the detected plate text. Finally, it generates a 8-second 16:9 MiniMax H3 reference-driven brand motion video with audio. It is best for logo-slogan identity reels and flat motion graphics openings, not for designing a new trademark, inventing slogans, or restoring official brand names from outside knowledge.","descCn":"本 Skill 面向已有 Logo、品牌名称图或含标语品牌视觉素材的用户，用于把静态品牌图转化为可复用的 H3 平面动态图形视频流程。用户提供一张 Logo 或品牌视觉图，也可以补充品牌说明，但最终视频中的所有可读文字必须来自图片本身或用户明确输入的品牌名 / 标语。流程先用 GPT Image 生成 16:9 横版品牌名称图，保留原图主题色与设计背景；再由当前模型直接读取生成后的横版图，并以该图中可见文字作为最终权威文本，只提取品牌名称与 slogan，按该图重新计算首字母、尾字母和前序文字锚点后改写固定 MiniMax H 动效 prompt；最后使用 MiniMax H3 生成 8 秒 16:9 带声音品牌动效视频。它适合品牌 Logo 标语动效、品牌片头和视觉识别展示，不适合从零设计商标、凭空补全 slogan 或依据外部知识还原官方品牌名。","tools":["question","hub_generate_image","hub_analyse_media","hub_generate_video","hub_save_file_to_session"],"skillType":"skill","sortWeight":700,"downloads":1583,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-14-21/video/1786713895599854613-1786713895590.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"sports-podcast-outline","version":"0.3.1","hash":"sha256:0f6219eca389e9b6a938e54588aa54dac80d9cd4b0a9270e6158b1c75637d683","summary":"Draft structured sports podcast episode outlines with segment timing, debate topics, hot takes, ad breaks, and guest interview prep.","summaryZh":"起草体育播客大纲，含时间轴、辩论点、观点、广告位、嘉宾问题。","description":"Draft a structured sports podcast episode outline as an expert sports podcast producer and content strategist. Input is the episode topic, target length, guest info (optional), and audience context; output is a segment-by-segment outline with timing, debate topics, hot takes, listener question integration, ad break placement, show flow, and guest interview prep. Use whenever the user wants to plan a sports podcast episode, structure segment timing, brainstorm debate points and hot takes, integrate listener questions, place ad breaks, or prep for a guest interview.","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":["sports podcast","podcast outline","episode outline","hot takes","debate segment","listener questions","guest interview prep","ad break placement","体育播客","播客大纲","节目大纲","犀利观点","嘉宾访谈"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"体育播客大纲","tagEn":"Audio Content / Post-Production","tagCn":"音频内容 / 后期制作","completeTagsEn":["Sound & Music / Post-production"],"completeTagsCn":["音频音乐 / 后期处理"],"descEn":"Sports Podcast Outline generates ready-to-record episode outlines as an expert sports podcast producer. Produces segment-level timing, debate points and hot takes with a clear angle, listener question integration, natural ad break placement, show flow optimization, and guest interview prep including opening questions and follow-up angles. Copy-paste templates the host can record straight against, grounded in current games, players, and storylines.","descCn":"体育播客大纲以资深体育播客制作人身份,产出可直接开录的节目大纲。包含段级时间轴、立场明确的辩论点与犀利观点、听众问题嵌入方式、自然过渡的广告位安排、节目节奏优化,以及嘉宾访谈的开场问题与追问角度。模板可直接复制使用,主持人对着大纲即可开录,内容锚定近期赛事、球员、故事线,拒绝空话套话。","tools":["question","hub_read","hub_write","hub_save_file_to_session"],"skillType":"skill","sortWeight":710,"downloads":1556,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"frontend-slides","version":"0.1.1","hash":"sha256:00c7546c619ab09e2cb190859f1b03b5c81bec681a847e3ba1f3d1ef35361399","summary":"Create distinctive, animation-rich HTML presentations from notes, content, or PPTX files with visual style previews and PDF/URL export.","summaryZh":"从零、笔记或 PPT 生成带动效 HTML 幻灯片，含风格预览与 PDF 导出。","description":"Create zero-dependency, animation-rich HTML presentations from scratch, rough notes, existing content, or PowerPoint files. Inputs are a topic/notes/PPTX/existing HTML deck; outputs are a distinctive, viewport-safe single-file HTML deck with optional PDF export or shareable URL. Uses visual style previews so users pick aesthetics by seeing, not by describing. Use whenever the user wants to build a presentation, convert PPT/PPTX to web, create slides for a talk/pitch, or enhance an existing HTML deck.","tags":["Education"],"tagsCn":["教育"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"前端幻灯","tagEn":"Platform Tooling","tagCn":"平台工具","completeTagsEn":["Education / Creative Generation"],"completeTagsCn":["教育 / 创作生成"],"descEn":"Frontend Slides creates distinctive, animation-rich HTML presentations that run entirely in the browser. Supports three modes: build from scratch, convert PPTX files, or enhance an existing HTML deck. Uses visual style previews so users pick aesthetics by seeing three concrete drafts rather than describing taste. Ships 12 curated style presets, a 34-template bold pack, viewport-safe fixed 16:9 stage, inline text editing, PDF export, and one-click Vercel deployment for sharing.","descCn":"前端幻灯用于生成完全在浏览器里运行、独具风格的动效 HTML 演示文稿。支持三种模式：从零创作、转换 PPTX 文件、增强已有 HTML deck。通过生成三份可视化风格预览让用户看着选审美，而不是说品味。内置 12 套精选风格预设、34 套 bold template pack、固定 16:9 视口安全舞台、浏览器内文本编辑、PDF 导出、以及一键部署到 Vercel 的分享方案。适合非设计师用户快速产出高质量演示文稿。","tools":[],"skillType":"skill","sortWeight":720,"downloads":1556,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"film-reference-prompt-writer","version":"1.0.0","hash":"sha256:0c063e8a17bce35f53a48191718c6c970ee2bdb89cc93e7c6ee8f05aef2c9a78","summary":"Turns film titles or movie stills into executable prompts by translating reference evidence into visual and camera rules.","summaryZh":"从电影名称、剧照或截图提炼可复用的视觉与镜头规则，生成可直接执行的图像或视频提示词。","description":"Analyze a film title, movie still, screenshot set, or visual reference to extract the observable film look, tone, lighting, color, framing, production design, camera movement, and director visual language, then turn a simple idea into an executable image or video prompt. Use this Skill specifically for reference-led prompt writing: it is not a film-style reference search tool or a generic camera-vocabulary guide. Not for full screenplays, long production plans, or automatic final media generation unless explicitly requested.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["电影参考提示词","剧照提示词","电影截图分析","电影质感迁移","reference-based film prompt","movie still prompt","film-look translation"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"电影质感提示词大师","tagEn":"Visual Production / Prompt Engineering","tagCn":"视觉创作 / 提示词工程","completeTagsEn":["Professional Film / Intent Brainstorming"],"completeTagsCn":["专业影视 / 意图脑爆"],"descEn":"For creators, directors, and AI visual users who want to translate an existing film reference into an executable prompt, this skill analyzes a film title, movie still, screenshot set, or short creative request and turns the reference evidence into practical image or video prompt language. It separates visible evidence, verified source facts, reasonable inference, and user-specified creative additions; inspects composition, lighting, tone, color, optics, production design, atmosphere, and movement; then writes a prompt with clear subject, framing, camera behavior, visual character, output requirements, and avoidances. It produces either a quick copyable prompt or a deeper reference analysis plus prompt. It is best for film-reference translation, screenshot breakdowns, and reference-based camera prompts; it is not a style-reference search catalog, a generic motion-vocabulary guide, a full screenplay tool, or a long production planner.","descCn":"面向想把电影参考转成可执行 AI 视觉语言的创作者、导演和提示词用户，解决只会写“电影感”“高级感”但模型难以执行的问题。用户可以输入电影名称、剧照截图、一组参考图，或一句人物、场景、情绪、运镜要求；技能会区分画面可见、资料确认、合理推断和用户指定内容，分析构图、光色、影调、镜头、焦点、美术、空气感与运镜，再输出快速可复制提示词，或完整视觉 DNA 分析加图像/视频提示词。它擅长电影质感迁移、摄影风格拆解和运镜提示词，不负责完整剧本、长流程制片规划，除非用户明确要求继续生成成片。","tools":[],"skillType":"skill","sortWeight":730,"downloads":1616,"coverUrl":"https://filecdn.minimax.chat/public/51d39501-404d-4fef-9c80-1ea93f4b374c.mp4","authorEn":"鲤鱼与鱼Ai","authorCn":"鲤鱼与鱼Ai","source":"community","installed":false,"updateAvailable":false},{"name":"podcast-marketing","version":"0.3.1","hash":"sha256:adcf1c7cecc0e96d74d0ed1288fd4d6e2f45e1148ced38b87c73143ee257714c","summary":"End-to-end podcast planning, production, and marketing playbook covering strategy, guest outreach, SEO show notes, growth, and monetization.","summaryZh":"播客全流程：策略、嘉宾外联、SEO shownote、增长与变现。","description":"Plan, produce, and market a podcast end-to-end as a growth channel. Covers concept and format selection, interview and solo episode structure, audio quality requirements, guest sourcing and outreach, SEO-optimized show notes, growth strategies from launch through scale, and revenue models. Input is a podcast concept or existing show; output is a full strategy document, guest outreach copy, or promotion checklist. Use whenever the user wants to launch a podcast, plan a podcast strategy, run guest outreach, write show notes, grow a podcast audience, or monetize an existing show.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["播客营销","播客策略","播客启动","嘉宾外联","shownote","podcast strategy","launch podcast","podcast marketing","guest outreach","show notes","grow podcast","podcast monetization"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"播客营销","tagEn":"Advertising / Idea & Concept","tagCn":"广告营销 / 创意构思","completeTagsEn":["Commercial Ad / Intent Brainstorming"],"completeTagsCn":["商业广告 / 意图脑爆"],"descEn":"Podcast Marketing is an end-to-end playbook for running a podcast as a marketing channel. Covers concept and format selection (interview vs solo), episode timing, audio quality baseline, guest sourcing and outreach templates, SEO-optimized show note structure, launch and growth strategies with per-episode promotion checklists, and monetization tiers from host-read ads and sponsorships to premium content and consulting. Outputs full strategy documents, outreach copy, and repurposing plans.","descCn":"Podcast Marketing 是把播客当营销渠道运营的端到端 playbook。覆盖概念与形式选择（访谈 vs 独白）、单集时间线、音频质量基线、嘉宾寻找与外联模板、SEO 优化的 shownote 结构、启动与增长策略（含单集推广 checklist），以及从 host-read 广告、赞助到付费内容和咨询的变现层级。产出完整策略文档、外联文案和内容复用计划。","tools":["question","hub_write","hub_save_file_to_session"],"skillType":"skill","sortWeight":740,"downloads":1514,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"anime-game-pv","version":"1.0.7","hash":"sha256:196248e7d0e47c0d192afdab73e145a9fddfd09b187c8dc2656628ff7d5c42a4","summary":"Create a complete anime or game character PV within 15 seconds, from visual lock to final video","summaryZh":"为二次元游戏与漫画角色制作十五秒内的完整宣传PV，从画风锁定到成片一站完成","description":"Create a complete anime, comic, or game PV up to 15 seconds from a story idea, character art, scene or style references, titles, or audio. Delivers one coherent film; not for long MVs or live action.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["anime PV","game PV","comic PV","character PV","story teaser","battle teaser","gacha promo","event PV"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"二次元游戏PV","tagEn":"Animation / Full Production","tagCn":"动画 / 成片制作","completeTagsEn":["Animation / Full Production"],"completeTagsCn":["动画 / 成片制作"],"descEn":"For anime, manga, and game creators who need a single polished character PV within 15 seconds. Takes character art, style references, layout KVs, action clips, or plain ideas as input, then walks through style authority confirmation, color contract, panel and logical-shot storyboarding, a master timeline, prompt compilation, and one full video render. Covers character reveals, awakening scenes, battle teasers, world showcases, and gacha or event promos, with strict identity and palette consistency across every shot.","descCn":"面向需要制作二次元角色宣传片的创作者，将角色立绘、三视图、场景或风格参考、排版视觉与创意描述转化为一条十五秒以内的完整PV成片。流程覆盖素材分析与身份锁定、画风来源确认、主题色与强调色合同、语义分镜与逻辑镜头规划、主时间线编排、提示词编译和整条视频生成，支持角色揭示、觉醒、战斗预告、世界观展示与抽卡活动宣传，全程保持角色身份、画风权威与色彩策略的一致性，最终交付单条可直接发布的宣传视频。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":750,"downloads":1431,"authorEn":"Official Design Team","authorCn":"官方设计团队","source":"official","installed":true,"updateAvailable":false,"installedVersion":"1.0.7"},{"name":"social-caption","version":"0.3.1","hash":"sha256:a306412248240e5c1c8b3d4e1a980b14b84a48f1bea758b553d46bf37ec870e0","summary":"Write native captions for Facebook, Instagram, TikTok, Pinterest, and YouTube — hook + payoff + CTA per platform's rules.","summaryZh":"为 FB、IG、TikTok、Pinterest、YouTube 写原生文案，hook + payoff + CTA 按各平台规则。","description":"Write captions for visual-first social media posts on Facebook, Instagram, TikTok, Pinterest, and YouTube. Input is the visual asset (photo, video, Reel, Short, pin) plus target platform, format, and goal; output is a platform-native caption with hook + payoff + CTA, hashtag rules applied, and length calibrated to each surface (Pinterest title vs description, YouTube title vs long-form description, Reel vs feed, Facebook vs IG). Use whenever the user wants an \"Instagram caption\", \"Reels caption\", \"TikTok caption\", \"Pinterest pin description\", \"Facebook caption\", \"YouTube title/description\", \"Shorts caption\", \"photo caption\", or shares an image/video and asks for the words that go with it. For text-first standalone posts see post-writer-sms; for carousels see carousel-writer-sms; for opening lines only see hook-writer-sms.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["Instagram caption","IG caption","Reels caption","TikTok caption","Pinterest description","pin caption","Facebook caption","YouTube title","YouTube description","Shorts caption","photo caption","caption this","社媒文案","IG 文案","Reels 文案","TikTok 文案","Pinterest 描述","YouTube 标题","配文"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"社媒文案","tagEn":"Platform Tooling","tagCn":"平台工具","completeTagsEn":["Commercial Ad / Post-production","E-Commerce / Post-production"],"completeTagsCn":["商业广告 / 后期处理","电商 / 后期处理"],"descEn":"Caption Writer produces platform-native captions for visual-first social posts. Ships explicit specs per surface: Facebook 40-500 chars, Instagram 125-char truncation + 3-10 hashtags, TikTok SEO-friendly under 150, Pinterest title ≤100 + description ≤500 with no hashtags, YouTube title 60-70 + description with chapters, Shorts under 150 with #shorts. Every caption follows the hook + payoff + CTA anatomy and comes with a pre-publish checklist.","descCn":"Caption Writer 为视觉优先社交贴生成平台原生文案。分平台明码规格：Facebook 40-500 字符、Instagram 125 字符 truncation 加 3-10 hashtag、TikTok 通常 150 字以内且带 SEO、Pinterest title ≤100 加 description ≤500 且不用 hashtag、YouTube title 60-70 加带 chapters 的长视频描述、Shorts 150 字以内且加 #shorts 标签。每条文案走 hook + payoff + CTA 结构，按 context 文件匹配用户声线，附发布前 checklist 覆盖 hashtag 规则、链接位置、alt text、chapters 等要点。","tools":["question","hub_read","hub_write","hub_save_file_to_session"],"skillType":"skill","sortWeight":760,"downloads":1450,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"brand-stream-mg","version":"1.0.2","hash":"sha256:d857e52e1dcba9938591733b33546cf3e96eefc94c78fe90fa32befeeaa8c2de","summary":"Generate a MiniMax-H3 brand stream MG video from one logo and theme, with IP-led motion and final logo fidelity.","summaryZh":"基于Logo与主题生成IP领航的品牌流线MG动效视频，适合新品发布预热与社媒传播。","description":"Create a 15-second brand-stream MG video from one logo and a brand name or theme. Develop selling points and an IP, icon, or energy lead, confirm the prompt, and keep textless logos free of text.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["brand-stream-mg","brand stream MG","brand MG","brand animation","logo animation","logo motion","brand video","MiniMax H3","15s brand film","fast brand MG"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"品牌流线 MG 动效","tagEn":"Commercial Ad / Creative Generation","tagCn":"商业广告 / 创作生成","completeTagsEn":["Commercial Ad / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成"],"descEn":"Creates a premium brand-stream MG video from one logo reference and a brand name, theme, or short concept. It extracts concise selling points, designs an IP, 3D icon, or energy-node lead, builds dual-neon stream motion, writes a 15-second production prompt, asks for confirmation, and generates the final video. MiniMax-H3 is the default; a user-selected other model requires capability checking first.","descCn":"面向品牌设计、产品发布和社媒投放场景，用户提供一张 Logo 参考图及品牌名、主题词或简短概念，Skill 会提炼英文或数字卖点，设计 IP、3D 图标或能量节点领航主角，组织双色霓虹流线动效，写入 15 秒制作提示词并在确认后生成视频。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败后按原因重试并切换可用模型。适合 Logo 动效、发布预热视频和品牌社媒短片，强调标志清晰与品牌一致。","tools":[],"skillType":"skill","sortWeight":770,"downloads":1469,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/7afd893c-12a6-4a63-82da-b9b908b391b4.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"cool-music-video","version":"1.0.2","hash":"sha256:2b6608c6af29ec03f9dcb41ff8d4bfff5ceaab25a3f85bd0056e624c9fc82f96","summary":"Create a 15-second retro collage, rap, or fashion performance music video as one continuous film","summaryZh":"制作十五秒以内的复古潮流拼贴、说唱或时尚表演型音乐短片，单条连续成片","description":"Create one 16:9 music-driven short up to 15 seconds from a song, performer reference, lyrics, or style direction. Best for collage, rap, and fashion performance; not full-song or multi-part MVs.","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":["short MV","music video","rap MV","fashion performance video","collage music video","15-second MV"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"潮流音乐短片","tagEn":"Sound & Music / Full Production","tagCn":"音频音乐 / 成片制作","completeTagsEn":["Sound & Music / Full Production"],"completeTagsCn":["音频音乐 / 成片制作"],"descEn":"For creators who want a single music-driven short within 15 seconds in 16:9. Takes a song, performer references, or a style direction as input, then confirms scope, selects character and music sources, designs eight to ten shots in one continuous space, plans dynamic typography and music structure, compiles a complete long-form video prompt, and dispatches one full render. Best for retro trend collages, rap performances, and fashion showcase videos where music rhythm drives every cut.","descCn":"面向需要音乐节奏主导短片的创作者，把歌曲、表演者参考或风格方向转化为一条十五秒以内、横屏画幅的音乐短片成片。流程覆盖范围确认、人物与音乐来源选择、八到十个镜头与连续空间设计、动态文字编排、音乐结构规划、完整长篇幅视频提示词编译与单条成片派发。适用于复古潮流拼贴、说唱表演与时尚走秀型音乐视频，以音乐节奏驱动每一次剪切，保持人物、空间与视觉风格在整条影片中的连续统一。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":780,"downloads":1409,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"1.0.2"},{"name":"wes-anderson-style-short-film-generator","version":"1.0.2","hash":"sha256:4145c0967d51bbb915406fc2bb9ea6d05c1be69b549cfee595d182009cd4bc33","summary":"Builds an original symmetry-led narrative short from a confirmed brief, references, storyboards, video segments, sound layers, and final assembly.","summaryZh":"将确认的故事与参考素材转成原创对称叙事短片，完成分镜、视频段、音频与合成。","description":"Create an original symmetry-led narrative, brand, or social short from a confirmed brief and references. Use for retro storybook visuals; not generic editing or copied films.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["Wes Anderson","Wes Anderson-style","symmetric composition","storybook short film","retro palette","对称构图","复古童话感","韦斯安德森","韦斯安德森风格短片","短片生成"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"韦斯·安德森美学短片生成器","tagEn":"Professional Film","tagCn":"专业影视","completeTagsEn":["Professional Film / Full Production"],"completeTagsCn":["专业影视 / 成片制作"],"descEn":"For creators who want an original narrative, brand, advertising, or social short built from centered symmetry, storybook staging, measured blocking, restrained retro design, and dry whimsy. The workflow confirms the brief, analyzes references, locks production specifications, creates visual direction and storyboards, generates mapped video segments, adds narration and sound, and assembles a validated film. MiniMax-H3 is the default video model; an explicitly selected alternative is checked for capability first, retried once on failure, then replaced with another available model. It does not copy specific films, shots, characters, dialogue, or an artist's exact signature.","descCn":"面向希望制作原创叙事短片、品牌片、广告或社交视频的创作者，核心是中心对称、童话式场面、精确调度、克制复古设计与干冷奇想。流程先确认故事简报和交付规格，再分析参考素材，锁定视觉方向，建立分镜与关键帧，生成有映射关系的视频段，补充旁白、音效和音乐，最后完成合成与校验。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。只做原创转译，不复制具体电影镜头、角色、对白或艺术家的精确个人签名。","tools":[],"skillType":"skill","sortWeight":790,"downloads":1458,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/82647356-8853-4892-b355-59fd34142187.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"typography-product-tvc","version":"1.0.2","hash":"sha256:8e759cc815ba8f3c3f4a3e627dded4b607e7a7b168325781e7d1e79ec7682ee5","summary":"Turns brand references into product multi-view images, a 3×3 storyboard, approved video segments, and a matched final TVC through five confirmation gates.","summaryZh":"以品牌参考图完成产品多视角、3×3分镜、分段视频与配乐成片，严格分步确认。","description":"Create a typography-led product TVC from brand references through product anchors, storyboard, approved video segments, and final assembly. Use for spatial text commercials and product hero films.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["typography product TVC","product hero film","spatial typography ad","产品贴字广告","产品TVC","品牌短片"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"产品贴字动画TVC","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成"],"descEn":"For brand teams and directors who need a typography-led product TVC from references and a brand name. The five-stage workflow clarifies inputs, creates product multi-view anchors, builds a 3×3 storyboard, generates approved video segments, and composites the final film with matched music and action sound. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once on failure, then replaced by another available model. It keeps product identity and text placement consistent and does not proceed when essential information is ambiguous.","descCn":"面向品牌方、广告创意与导演团队，把参考图和品牌名转成以贴字动效为核心的产品 TVC。流程严格分五步：先澄清素材与规格，生成并确认产品多视角锚定图，再生成并确认 3×3 分镜，按分镜逐段生成视频并确认，最后合成片段、匹配 BGM 与动作音效。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。全程保持产品身份、文字位置和节奏一致，关键信息不清晰时先询问。","tools":[],"skillType":"skill","sortWeight":800,"downloads":1463,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/94816a0d-fb3e-47ce-9004-ec9a07fb6514.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"video-deconstruct","version":"1.0.2","hash":"sha256:0f510aa55d720826afedbbdc5033d7b4e5acd2280dd2fc92db61678d32f8184c","summary":"Deconstruct reference videos shot by shot and compile verifiable evidence into a reproducible video prompt","summaryZh":"逐镜拆解参考视频、关键帧或详细描述，编译为可复刻的最终视频提示词与编辑计划","description":"Deconstruct video, keyframes, screenshots, or a description into evidence on shots, motion, sound, and continuity, and deliver a recreation prompt and edit plan. Not for critique or plain editing.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["video deconstruction","recreate this video","reverse engineer video prompt","shot-by-shot analysis","video prompt reconstruction","structure-preserving remake"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"视频拆解复刻","tagEn":"Creative & Experimental / Planning","tagCn":"创意实验 / 计划制定","completeTagsEn":["Creative & Experimental / Planning"],"completeTagsCn":["创意实验 / 计划制定"],"descEn":"For creators who want to recreate or reverse engineer a reference video. Takes a video file, keyframes, or a detailed written description as input, then extracts verifiable evidence on shots, camera motion, subject action, sound, and continuity, and compiles it into a reproducible final video prompt plus an editing plan. Useful for shot-by-shot analysis, structure-preserving remakes, and prompt reconstruction workflows where fidelity to the original pacing and camera language matters.","descCn":"面向需要复刻或反推参考视频的创作者，把用户提供的视频、关键帧或详细文字描述拆解为可验证的镜头、运动、声音与连续性证据，并编译成可复刻的最终视频提示词与编辑计划。适用于逐镜拆解分析、保持原片结构的重制版本以及提示词反推等工作流。拆解结果覆盖镜头顺序、摄影机路径、主体动作、转场逻辑与声音事件，保证复刻版本在节奏、运镜语言与叙事结构上与原片保持一致，最终交付可直接执行的生成方案。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":810,"downloads":1341,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"1.0.2"},{"name":"live-sketch-motion","version":"1.0.1","hash":"sha256:fd240c66377dc964161e1ef08503520393be633b537074215c6f1677ee6828f0","summary":"Turns real photos, videos, or text scenes into sketch-motion clips where a marker reveals a moving subject as hand-drawn art.","summaryZh":"让真实照片先自然动起来，再由笔尖把运动主体显影成手绘短片，保留实景空间与生命感。","description":"Create live-action sketch-motion clips from a real photo, reference video, or text scene. Use for live sketch, live-action sketch motion, marker reveal, photo-to-sketch-motion, doodle motion, or moving subjects progressively becoming hand-drawn. The workflow uses a realistic preview for text scenes, natural pre-drawing motion, marker-tip synchronized reveal, subjects that never freeze, colored-pencil doodle texture, and subtle live-action background motion. Not for static stickers, subtitles, or unrelated VFX.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["Live Sketch Motion","live sketch","live-action sketch motion","sketch motion","marker reveal","marker sketch","photo-to-sketch-motion","doodle motion"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"实景动态手绘显影","tagEn":"Creative & Experimental","tagCn":"创意实验","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"Creates short live-action sketch-motion clips from a real photo, reference video, or text scene. Text-only input first receives a realistic preview for spatial confirmation; the approved workflow then uses natural pre-drawing motion, marker-tip reveal, living subjects, restrained hand framing, colored-pencil texture, and subtle background motion. MiniMax-H3 is the default video model, while another model requires explicit user selection and a capability check.","descCn":"面向想把实景照片、参考视频或文字场景做成短视频手绘显影效果的创作者。纯文字输入会先生成真实预览，用于确认空间、主体位置和遮挡关系，再产出制作方案与详细提示词，确认后生成视频。规则强调手绘前先自然视频化、笔尖同步显影、主体转换前中后持续运动、手从边缘克制入画、彩铅或蜡笔排线质感、背景生活化微动与主体完整可见。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力再执行，失败后切换可用模型。","tools":["question","hub_generate_image","hub_generate_video","hub_analyse_media","hub_save_file_to_session"],"skillType":"skill","sortWeight":820,"downloads":1348,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/51170a8f-eb5d-4285-b0fe-3922c6577959.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"h3-visual-design","version":"1.0.1","hash":"sha256:9099c1e97589d1c2ce6a4bd1d59ba8f5c5525de2207a5367b7ad26b5790ee1f5","summary":"Single-technique motion design covering kinetic typography, subject tracking graphics, and hand-drawn live-action fusion","summaryZh":"单点动态视觉设计入口，覆盖动态字体包装、主体追踪图形与手绘实拍融合三类技法","description":"Create H3 motion-design video using kinetic typography, tracking graphics, or live-action and hand-drawn interaction. Requires a visible moving technique; not for static design or generic animation.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["H3 motion design","kinetic typography","motion typography","tracking graphics","TouchDesigner look","hand-drawn live action","motion packaging"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"动态视觉设计","tagEn":"Creative & Experimental / Creative Generation","tagCn":"创意实验 / 创作生成","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"For creators who need one specific motion technique executed well in a short video. Supports three professional workflows: kinetic typography and spatial packaging for fonts, logos, titles, and captions; tracking frames, topology lines, digital IDs, and partial-negative effects that follow a moving subject; and continuous morphing and physical interaction between live footage and 2D doodle, crayon, or chalk animation. Each route handles setup confirmation, material constraints, executable video prompts, and generation or delivery.","descCn":"面向需要特定动态技法的创作者，在一条短视频中精确执行单点视觉设计。支持三类专业工作流：字体、标志、标题、口播字幕或图形的动态版式与空间包装；追踪框、拓扑线、数字编号、局部负片等跟随主体运动的视觉表现；真人实拍与二维手绘、涂鸦、蜡笔或粉笔动画之间的连续变形、物理接触和互动。各路线覆盖设置确认、素材约束、可执行视频提示词编译，以及对应的视频生成与交付，保证技法表达准确、画面运动可控。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":830,"downloads":1312,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"1.0.1"},{"name":"anime-pv-maker","version":"0.1.2","hash":"sha256:9b6b34c1a8a467a4f75e2cba5fdfdc85c8d2b80949e5d9444006faf3a85096e5","summary":"Turn short audio and visual references into rhythm-synced anime PV concepts and finished clips.","summaryZh":"将短音频与视觉参考转化为节奏同步的二次元PV创作方案并输出完整成片","description":"15秒以内音频/参考视频驱动的完整二次元 PV、MV、一图流、文字动画PV、动态设计MG 制作 Skill。启动时先用卡片询问用户要制作 MV、一图流/文字动画PV、动态设计MG 或其他类型，并确认可提供素材：纯音频+想法、纯音频+视频参考、图片/立绘+分镜图参考、参考动画等。随后分析音频或参考视频音轨，判断歌词、鼓点、弦乐、节奏点与情绪段落；若有歌词则索要准确歌词并匹配节奏，纯音乐则自动处理鼓点弦乐。参考图/视频会自动解析美术风格并反解 prompt，立绘作为分镜主角元素，分镜图参考作为MG/PV单帧简略演示。视频生成模型锁死 MiniMax H3 最高精度，输出9/16宫格分镜、长提示词、H3片段和完整卡点成片。强调曲线有力、合成精细、排版紧凑、动画敏捷、最高规格最高质量。","tags":["Animation","Sound & Music"],"tagsCn":["动画","音频音乐"],"creator":"","triggerWords":["PV","二次元PV","MV","一图流","文字动画PV","动态设计MG","角色PV","音频做PV","音乐PV","卡点PV","参考动画PV","完整PV"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"二次元PV","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Planning","Animation / Creative Generation","Sound & Music / Creative Generation"],"completeTagsCn":["动画 / 计划制定","动画 / 创作生成","音频音乐 / 创作生成"],"descEn":"Turns audio, reference videos, character art, stills, or storyboards under 15 seconds into anime PV, MV, text-animation PV, and motion-design MG workflows. It confirms format and available materials, analyzes lyrics, beats, strings, emotion, and reference style, proposes 9/16-panel storyboards, writes high-density H3 prompts, generates high-quality clips, and assembles a beat-synced final video while preserving confirmed character, layout, lyric, and style decisions.","descCn":"将15秒以内的音频、参考视频、二次元立绘、静帧或分镜转化为二次元PV、MV、文字动画PV和动态设计MG工作流及完整成片。先通过卡片确认制作类型与可用素材，再分析歌词、鼓点、弦乐、情绪段落和参考风格，规划9/16宫格分镜，生成高密度H3长提示词与高精度片段，并按节拍完成镜头、文字、转场和音频合成。流程会保留用户确认的角色、构图、歌词、画风和分镜结构，并在生成前提供必要的方向确认。","tools":[],"skillType":"skill","sortWeight":840,"downloads":1340,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"ip-toy-grid-motion","version":"1.0.2","hash":"sha256:bcb9f57220fb7082b8afadc6410d65ace19414131fde87c6f073d65087493b41","summary":"Generates a vertical 3D designer-toy six-panel motion video from a user-uploaded character image.","summaryZh":"基于用户上传的角色形象，生成3D潮玩六宫格竖版动态视频，适合IP提案与社媒展示。","description":"Converts a user-uploaded IP character image into a 9:16 designer-toy six-panel motion poster. Use when the user wants an IP character, mascot, anime figure, brand character, or original character transformed into a hyper-realistic 3D vinyl blind-box toy collage, then animated into a short vertical grid video. This Skill preserves the original Chinese prompt templates stored in references and only performs minimal replacements for character identity, props, or conflicting colors. Trigger words: IP toy, toy six-panel poster, designer-toy motion poster, IP-to-3D toy, toy figure grid video, designer toy motion poster.","tags":["Creative & Experimental","E-Commerce"],"tagsCn":["创意实验","电商"],"creator":"","triggerWords":["IP toy","toy six-panel poster","designer-toy motion poster","IP-to-3D toy","designer toy","toy figure grid","motion poster"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"IP潮玩六宫格动态海报","tagEn":"Creative & Experimental","tagCn":"创意实验","completeTagsEn":["Creative & Experimental / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成","电商 / 创作生成"],"descEn":"Converts one uploaded IP character image into a 9:16 designer-toy six-panel poster and a six-second motion version. Panel 4 opens full screen, returns to the grid, and the full layout animates subtly while identity, props, coverage, colors, and panel structure remain consistent. MiniMax-H3 is the default video model; another model requires explicit user selection and capability checking.","descCn":"本 Skill 面向需要把单个 IP 角色、吉祥物、二次元角色、品牌角色或原创形象转化为短视频视觉资产的创作者。用户提供一张角色图，也可补充角色名、音频或文字需求。流程先生成 9:16 黑底不规则圆角六宫格静态海报，再生成 6 秒竖版动态海报：宫格4先全屏开场，再缩小归位并让其他宫格飞入，最后六宫格稳定微动。图像和视频阶段都保持角色身份、身体覆盖状态、道具、背景色逻辑、宫格布局和运动结构。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力再执行。","tools":["hub_analyse_media","hub_read","hub_generate_image","hub_generate_video","hub_canvas_group_recent_outputs"],"skillType":"skill","sortWeight":850,"downloads":1337,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/59c62dd8-8b28-4b31-b559-99e717119928.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"podcast-studio","version":"0.3.1","hash":"sha256:87008b6e0c918c36b29622a6e4a45d47841b7f2dd6ce9727939f669d0411150f","summary":"Turn written content into a podcast script with narration timing, music cues, and sound effect direction.","summaryZh":"把书面内容改写成播客脚本，附带朗读时长、音乐 cue 和音效指令。","description":"Turn written content into a podcast script with audio timing, music cues, sound effects, and production notes. Input is any long-form text (article, blog post, essay, briefing); output is a structured markdown script that a producer can hand to a voice actor and audio engineer. Use whenever the user wants to convert an article to a podcast, draft show notes with music cues, plan intro/outro music around Tone.js or Howler.js, or spec out sound design for a spoken piece.","tags":["Sound & Music"],"tagsCn":["音频音乐"],"creator":"","triggerWords":["podcast script","article to podcast","podcast production","show notes","music cues","sound design","Tone.js","Howler.js","播客脚本","播客制作","文章转播客","音乐 cue","声音设计"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"播客工作室","tagEn":"Audio Content / Post-Production","tagCn":"音频内容 / 后期制作","completeTagsEn":["Sound & Music / Post-production"],"completeTagsCn":["音频音乐 / 后期处理"],"descEn":"Podcast Studio transforms any long-form written piece into a producer-ready podcast script. It segments the text into cold open, intro, chapters, and outro, attaches audio direction (music beds, transitions, SFX), and estimates spoken duration for each section. Music cue guidance assumes a browser-based delivery stack using Tone.js for synthesized beds and Howler.js for discrete sound effects. Ideal for turning articles, essays, or briefings into podcasts, drafting show notes with audio cues, and specifying sound design for spoken content.","descCn":"Podcast Studio 把任意长文转成可直接进棚的播客脚本。按 cold open、片头、章节、结尾拆分文本，为每段附上音频指令（音乐 bed、转场、SFX），并给出朗读时长估算。音乐 cue 假设浏览器端交付栈：Tone.js 做音乐 bed，Howler.js 触发离散音效。适合把文章、专栏、简报改成播客节目，起草带音频 cue 的 shownote，或给口播内容规划声音设计。","tools":["question","hub_read","hub_write","hub_save_file_to_session"],"skillType":"skill","sortWeight":860,"downloads":1298,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"subtitle-correction","version":"0.1.1","hash":"sha256:563888a8feda05cb755d39f0d02500e41c25ffc2fc1eaf71900f199b0049d7af","summary":"Correct speech recognition errors in SRT subtitles while strictly preserving timeline and numbering.","summaryZh":"修正 SRT 字幕语音识别错误，严格保留时间轴和编号。","description":"Correct speech-recognition errors in subtitle files (`.srt`) while strictly preserving timeline information and subtitle numbering. Interactively asks the user for domain-specific terminology first (frameworks, brands, technical terms), then applies pattern-aware corrections tuned for that domain — Chinese phonetic confusions, technical term misrecognitions, English-Chinese mixed content, and code identifier spelling. Ships a `subtitle_tool.py` validator and diff reporter for review. Use whenever the user uploads a `.srt` file and asks to correct / fix / proofread subtitles, especially for programming tutorials, AI/ML courses, or any content with domain-specific jargon.","tags":["Platform Tooling"],"tagsCn":["平台工具"],"creator":"","triggerWords":[],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"字幕修正","tagEn":"Audio Content / Post-Production","tagCn":"音频内容 / 后期制作","completeTagsEn":["Platform Tooling"],"completeTagsCn":["平台工具"],"descEn":"Subtitle Correction fixes speech-recognition errors in SRT files while strictly preserving all timestamps and subtitle numbering. Interactively asks the user for domain-specific terminology first — proper nouns, technical terms, framework names — then applies domain-aware corrections tuned for that vocabulary. Covers Chinese phonetic confusions, technical term misrecognitions, English-Chinese mixed content, and code identifier spelling. Ships a subtitle_tool.py validator, diff viewer, and HTML report generator for structured review before finalizing changes.","descCn":"字幕修正修 SRT 文件里的语音识别错误，严格保留所有时间戳和字幕编号。会先向用户索取领域术语——专有名词、专业术语、框架名——然后按该词汇领域调优做修正。覆盖中文谐音混淆、专业术语误识、中英文混排、代码标识符拼写。附带 subtitle_tool.py 校验器、diff 查看器和 HTML 报告生成器，方便修改定稿前做结构化 review。适合技术教程、AI/ML 课程等带专业术语的字幕修正。","tools":[],"skillType":"skill","sortWeight":870,"downloads":1304,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":false,"updateAvailable":false},{"name":"ui-motion","version":"1.0.1","hash":"sha256:00eb2e1b35d54f6e5f4b1132307997911bb3b12a8114b5f202f0567d7b918dda","summary":"Turn real brand visuals into UI motion graphics, app interface animations, and SaaS product demo videos","summaryZh":"基于真实品牌视觉制作界面动效、应用界面动画与产品演示视频的端到端入口","description":"Create UI motion, app-interface animation, or a SaaS product demo from real brand visuals, exact text, and interaction goals. Delivers a brand-consistent video; not static design or live-action ads.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["UI motion","app interface animation","SaaS hero animation","product interface demo","cursor interaction","dashboard animation"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"界面动效演示","tagEn":"Creative & Experimental / Full Production","tagCn":"创意实验 / 成片制作","completeTagsEn":["Creative & Experimental / Full Production"],"completeTagsCn":["创意实验 / 成片制作"],"descEn":"For product and brand teams that need UI motion graphics, app interface animations, SaaS hero animations, or interface demos built from real brand visuals. Takes app screenshots, web pages, dashboards, or written brand descriptions as input, then runs brand visual analysis, builds a brand profile, selects inspiration and motion style, plans a timecoded storyboard, compiles continuous-transition prompts, generates video, checks brand consistency, and delivers the final MP4. Sequences over 15 seconds use true last-frame continuation with a single full BGM pass.","descCn":"面向产品与品牌团队，基于真实品牌视觉制作界面动效、应用界面动画、产品首屏动效或界面演示视频。输入应用截图、网页、数据面板或明确的品牌文字描述后，流程依次完成品牌视觉分析、品牌档案建立、灵感与动效风格选择、带时间码的分镜规划、连续转场提示词编译、视频生成、品牌一致性质检与最终成片交付。十五秒以内生成单条；超过十五秒时使用真实尾帧连续生成并顺序拼接，最后统一生成一次完整背景音乐。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":880,"downloads":1206,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"1.0.1"},{"name":"niulai-movie-style","version":"1.0.0","hash":"sha256:4691c655d5d0f9af947bc19d1adf9fc204ee478fd87311bc6bb02f477f8915aa","summary":"Guide users from uploaded photos to deliberately failed primitive low-poly images and optional short videos.","summaryZh":"引导用户上传照片，生成故意失败的原始低模图像，并可继续延展成笨拙运镜短视频。","description":"Turn a supplied photo into awkward primitive low-poly 3D. Use for Niulai movie style, primitive folk CGI, bootleg CGI, early game imagery, crude characters, or awkward motion/camera. Preserve subject count, layout, action, camera, scene type, props, and colors while degrading geometry, proportions, rigging, collisions, textures, lighting, rendering, and motion. Requires an image unless a new scene is requested; not for pixelation, VHS, horror distortion, low-poly art, or cinematic animation.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["niulai movie style","niulai camera movement","protagonist motion","reverse viral low quality","crude character low-poly","early 3D game screenshot","primitive folk CGI","bootleg CGI","failed low-poly"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"牛来电影风格","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For creators who want an intentionally cheap and awkward early-3D reconstruction rather than polished low-poly art. The Skill first guides users to upload a photo or source image unless they explicitly ask for a new scene, then extracts broad anchors such as subject count, layout, action, camera, scene type, props, and large color blocks. It then rebuilds the image with extremely sparse geometry, broken proportions, stiff rigging, local clipping, crude faces, tiny repeated diffuse textures, reused assets, naive lighting, and cheap rendering. The deliverable is an edited image, followed by a lightweight prompt asking whether to continue into a short video; it can also provide a video-ready prompt or production prompt when requested. It is best for primitive folk CGI, bootleg CGI, and reverse-viral low-quality aesthetics; it does not create simple pixelation, VHS filters, horror deformation, ordinary polished low-poly illustration, or smooth cinematic animation.","descCn":"面向想把源图转成“制作能力不足”的早期3D效果的创作者，而不是做精致低模艺术。Skill 会先引导用户上传照片或源图；除非用户明确要求生成新场景，否则先基于图片提取主体数量、布局、动作、镜头、场景类型、主要道具和大色块等锚点，再用极低面数、失真比例、僵硬绑定、局部穿模、粗糙脸部、低分辨率重复贴图、复用资产、朴素灯光和廉价渲染进行重建。最终先交付编辑图，并轻量询问是否继续生成短视频；也可在用户要求时交付视频可用提示词或生产提示词。适合 primitive folk CGI、bootleg CGI 和反向出圈低画质审美，不用于普通像素化、VHS 滤镜、恐怖变形、常规精致低模插画或顺滑电影动画。","tools":["question","hub_analyse_media","hub_generate_image","hub_generate_video","hub_save_file_to_session"],"skillType":"skill","sortWeight":890,"downloads":1212,"coverUrl":"https://cdn.hailuoai.com/pre/2026-08-18-17/video/1787046735278572549-1787046735272.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"cinematic-title-sequence","version":"1.0.1","hash":"sha256:11b15bab51171f116791bb6fd6339967cea52096588ec5180a948b7a94019d56","summary":"Create film or series opening title sequences and 15-second concept teasers ending on the title reveal","summaryZh":"制作电影或剧集片头标题序列与十五秒影视概念预告，以片名或卡司落板收束","description":"Create one 15-second film or series title sequence, cast sequence, or concept teaser from confirmed story, exact text, visual references, typography, and sound. Not for full trailers or end credits.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["title sequence","opening titles","opening credits","cast sequence","film teaser","series teaser","concept teaser"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"影视片头序列","tagEn":"Professional Film / Full Production","tagCn":"专业影视 / 成片制作","completeTagsEn":["Professional Film / Full Production"],"completeTagsCn":["专业影视 / 成片制作"],"descEn":"For filmmakers and series creators who need an opening title sequence, credits sequence, or a single 15-second concept teaser. Takes stills, posters, title text, cast names, or a story premise as input, then routes scope, selects the master style and typography system, confirms exact on-screen text, compiles shots and transitions into a long-form video prompt, and generates the final film with synchronized music. The narrative uses the film world, characters, and conflicts, always closing on the title or cast reveal.","descCn":"面向需要影视片头或概念预告的创作者，把剧照、海报、片名、卡司文字或故事设定转化为一条完整的片头标题序列或十五秒影视概念预告。流程覆盖成片范围分流、主风格与文字系统选择、参考素材约束、原创影片方案与准确文字确认、镜头与转场编译、长篇幅视频提示词生成、同步音乐与最终成片交付。叙事以影片世界、人物行动、关系与冲突推进，最终以片名、卡司或预告落板收束，保证文字准确与画面连续。","tools":[],"skillType":"skill","badges":["silent-install"],"sortWeight":900,"downloads":1069,"authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official","installed":true,"updateAvailable":false,"installedVersion":"1.0.1"},{"name":"browser-game-demo-maker","version":"1.0.2","hash":"sha256:196bf4315dc49622cd305419458f2809e4790867cd96fe33384e8c74600376de","summary":"Creates a realistic browser localhost game-demo screenshot and a matching action video with the same browser frame and character continuity.","summaryZh":"面向本地浏览器游戏演示，先产出真实截图锚点，再生成保持同框的动作视频。","description":"Create a Chrome screenshot of a localhost HTML5 game, then a matching video with the same browser frame and character continuity. Uses supplied prompt files; not for posters or generic motion.","tags":["Creative & Experimental"],"tagsCn":["创意实验"],"creator":"","triggerWords":["-","browser","screenshot"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"浏览器游戏 Demo 制作","tagEn":"Creative & Experimental","tagCn":"创意实验","completeTagsEn":["Creative & Experimental / Creative Generation"],"completeTagsCn":["创意实验 / 创作生成"],"descEn":"This Skill is for users who want a realistic desktop browser capture that looks like a local localhost HTML5 game demo, followed by a matching video that preserves the same browser frame and only animates inside the game canvas. It takes a reference character image, exact browser UI text, local URL details, control labels, and action sequence notes, then produces a screenshot anchor and a continuity-safe motion version. Use it for browser-game feel tests, retro arcade UI demos, and screenshot-to-video continuity when the browser chrome must stay visible. It is not meant for poster layouts, static UI mockups, generic motion graphics, or unrelated gameplay editing.","descCn":"这个 Skill 面向“真实桌面浏览器窗口 + 本地 localhost HTML5 游戏 Demo”的连续制作需求。用户提供参考角色图、浏览器界面文案、本地 URL、按键说明和动作顺序后，先生成一张保留完整浏览器外壳的截图锚点，再基于同一框架生成只在画布内变化的视频。适合浏览器游戏手感测试、复古街机 UI 演示和截图到视频的连续交付；不适合海报、静态 UI Mockup、纯动效包装或与本地网页游戏无关的剪辑任务，强调画面连续与角色一致。","tools":[],"skillType":"skill","sortWeight":910,"downloads":1058,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/60ce975d-7331-4650-b4ca-2b3c5c3ae4ac.mp4","authorEn":"Victor M","authorCn":"Victor M","source":"community","installed":false,"updateAvailable":false},{"name":"handdrawn-product-poster-motion","version":"1.0.4","hash":"sha256:bb2c693c630d6fa0c908c4e5e326faf7635e0cc9002a75832e07cd5841fbc0ba","summary":"Transforms a small product photo into a hand-drawn POP poster and a seven-second making-of animation through confirmed title, stickers, and footer copy.","summaryZh":"把小物产品图制作成手绘POP海报与7秒制作动画，分步确认标题、贴纸和底部文案。","description":"Create a hand-drawn POP product poster and short making-of animation from a small product image. Use for drinks, cosmetics, gadgets, snacks, stationery, or toys; not large products or long ads.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["product poster motion","handdrawn product poster","doodle product ad","sticker poster animation","小物产品海报","手绘海报动画","产品贴纸动画","POP产品广告"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"小物产品手绘海报动画","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Full Production","E-Commerce / Full Production"],"completeTagsCn":["商业广告 / 成片制作","电商 / 成片制作"],"descEn":"For creators presenting drinks, cosmetics, phones, earbuds, snacks, stationery, toys, and other hand-holdable products. The Skill confirms the product title, sticker strategy, and optional footer copy in separate checkpoints, loads bundled poster references, creates a vertical POP poster combining realistic product photography with marker doodles and illustrated hands, then animates the making process. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once when it fails, and then replaced by another available model. Not for large products, talking-head commerce, UI demos, luxury cinematic campaigns, or long narrative ads.","descCn":"面向需要把饮品、美妆、手机、耳机、零食、文具、潮玩等小物快速包装成年轻视觉的创作者。Skill 按检查点确认产品标题、贴纸策略和可选底部小字，调用内置海报提示词生成真实产品摄影融合马克笔涂鸦、插画双手与彩色贴纸的竖版 POP 海报，再把制作过程延展为 7 秒固定镜头动画。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。适合小物展示，不适合大型产品、真人带货、UI 演示、高奢电影广告或长剧情。","tools":[],"skillType":"skill","sortWeight":920,"downloads":1053,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/6f9c6da8-628e-424c-8367-480041835658.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"material-propagation-poster","version":"1.0.2","hash":"sha256:e32504050a2d189f64febcfb985a61251c2a26681c797d6b0084a5d89f391372","summary":"Turns a subject visual into an approved material-journey poster with protected layout, controlled motion, and social variants.","summaryZh":"从主体视觉确认关键画面，制作保护版式的材质蔓延动态海报并输出社媒变体。","description":"Turn a subject visual and optional texture reference into a still key art, restrained material-motion poster, and social variants. Use for visible surface journeys, not ordinary poster design.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["material propagation","material journey","surface takeover","ink spread poster","coating animation","patina growth","crystal growth poster","dynamic material poster","surface growth poster","material spread"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"材质蔓延动态海报","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成"],"descEn":"For designers and creators who want a subject image, poster, or product visual to become a controlled material-propagation package. The workflow identifies the visual anchor and protected regions, defines the material origin, route, timing, stopping rule, and loop, confirms a still key art, and then creates a restrained motion poster with social-format variants. It preserves identity, geometry, typography, lighting, and layout. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once on failure, then replaced by another available model. Not for ordinary static posters, subtitle editing, generic image-to-video, or standard cutting.","descCn":"面向设计师与动态海报创作者，把主体图、海报或产品视觉转成可控的材质蔓延制作包。流程先识别视觉锚点与保护区域，明确材质起点、传播路径、时间节奏、停止规则和循环方式，确认静态关键画面后，再生成克制的动态海报与社媒比例变体。整个过程保留主体身份、几何结构、文字、光影和版式。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。不用于普通静态海报、纯字幕、通用图生视频或常规剪辑。","tools":[],"skillType":"skill","sortWeight":930,"downloads":1046,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/3052949b-b42f-42bb-8d39-0380855b300b.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"mime-object-video","version":"1.0.3","hash":"sha256:8a8e2ba41028562aa3f5ac35435200a3c6d4761e9609bc7ead371f7a302b48b6","summary":"Adds a referenced object to a mime-acting video while preserving the original person, scene, camera, timing, and performance.","summaryZh":"把参考物品自然加入无实物表演视频，同时保持人物、场景、镜头与动作不变。","description":"Add one to three referenced objects to a mime-acting video while preserving the person, scene, camera, timing, and performance. Use for no-prop acting enhancement and object insertion.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["mime acting","no-prop acting","imaginary object","prop insertion","object video edit","hand prop","wearable prop","无实物表演","无道具表演","物品增强","道具植入","把物品加到视频里","手上出现物品","戴上参考物品"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"珠宝佩戴视频生成","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成","电商 / 创作生成"],"descEn":"For creators who film a person acting without a physical prop and want one to three referenced objects inserted on a chosen body area. The workflow analyzes source motion, visibility, lighting, camera movement, and object structure, creates an optional placement anchor, then generates an edit that preserves the original person, scene, timing, and performance while maintaining plausible scale, perspective, contact, shadow, occlusion, and continuity. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once on failure, then replaced by another available model. It is not for full scene redesign, product ads, or multi-shot production.","descCn":"面向拍摄无实物表演素材、希望把 1–3 件参考物品自然加入指定身体部位的创作者。流程先分析源视频动作、目标部位可见性、光线、镜头运动和物品结构，必要时生成物品放置锚点，再进行保留优先的视频编辑。最终应保持原人物、服装、场景、镜头、节奏和动作不变，并让物品具备合理比例、透视、接触、阴影、遮挡和运动连续性。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。不用于完整场景重设计、产品广告或多镜头成片。","tools":[],"skillType":"skill","sortWeight":940,"downloads":1046,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/9a9f6ff7-1ae4-457a-b1b6-968889d017d2.mp4","authorEn":"地质大学博士说AI","authorCn":"地质大学博士说AI","source":"community","installed":false,"updateAvailable":false},{"name":"logo-creative-material-motion-lab","version":"1.0.3","hash":"sha256:5dbaddf4c71335c6629e43f3033594eec84fdcc77119884d27aae0e3711cd688","summary":"Builds a logo motion package with paired 3D bases, three material directions, and horizontal and vertical video variants.","summaryZh":"以Logo和参考图生成3D基底、三种材质方向及横竖版品牌动效视频，适合TVC与社媒切条。","description":"Create a logo motion package from a logo, brand tone, and references, covering a 3D base, material directions, and horizontal or vertical video variants. Use for brand reveals and TVC openers.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["logo animation","logo reveal","3D logo","logo motion","片头","片尾","LOGO动效","立体logo","创意材质动效"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"Logo创意材质动效实验室","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成","电商 / 创作生成"],"descEn":"For designers, brand teams, and motion creators who need a structured logo motion package. The workflow pairs one logo with one reference image to create 3D material bases, writes three distinct directions with horizontal and vertical prompts, then generates six short videos with MiniMax-H3 by default. If the user explicitly chooses another model, its capabilities are checked first; a failure is retried once for the diagnosed cause, then moved to another available model. The package is suited to brand reveals, TVC openers, and social cutdowns, not unrelated editing or long-form story planning.","descCn":"适用于设计师、品牌团队和动效创作者的结构化 Logo 动效流程。用户提供 Logo、品牌名、品牌调性和可选参考图后，Skill 坚持一张 Logo 配一张参考图，逐一生成 3D 材质基底，整理三条互相区分的材质动效方向，每条包含横版与竖版提示词，最后默认使用 MiniMax-H3 生成 6 条短视频。用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。适合品牌开场、TVC 和社媒切条，不负责无关剪辑或长篇叙事。","tools":[],"skillType":"skill","sortWeight":950,"downloads":1053,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/033fd37b-8953-4928-9d93-69c22da8e43d.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"brand-ip-asset-package-design","version":"1.0.2","hash":"sha256:2046b00ae4c92e0a0d126975b7b224209a876e0b7fde0bbce9b4c2bbd143dfac","summary":"Builds a reusable IP asset system from a sketch or brief through staged character, expression, outfit, poster, and merch modules.","summaryZh":"从IP草稿或简报分步扩展角色、表情、服装、海报与文创周边，形成完整可复用资产包。","description":"Build a reusable IP character asset system from a sketch, reference, or brief, covering characters, expressions, outfits, posters, and merchandise. Use for staged IP development.","tags":["Commercial Ad"],"tagsCn":["商业广告"],"creator":"","triggerWords":["IP物料","IP角色","表情包","换装","海报","文创周边","潮玩","手办"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"品牌IP资产包设计","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成"],"descEn":"For users who want one IP character expanded into a reusable asset system rather than a single image. The workflow locks identity from a sketch, reference, or text brief, then proceeds through 3D character, expression pack, outfit variants, themed posters, and merchandise modules. Each module is a checkpoint: show the result, collect changes, and ask for confirmation before continuing. Image and motion outputs use MiniMax-H3 as the default video path; an explicitly selected alternative is capability-checked, retried once on failure, then replaced by another available model. Best for IP development, toy design, and child-friendly merchandise, not long-form video or one-click batching.","descCn":"适用于想把一个 IP 角色扩展成可复用物料系统的用户，而不是只生成一张图。用户可提供草稿图、参考图或文字设想，流程先锁定角色身份，再依次制作 3D 形象、表情包、不同服装主体、主题场景海报和文创周边。每个模块都是独立检查点：展示结果、收集修改并确认后才进入下一步。涉及视频时默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。适合 IP 角色开发、潮玩和儿童向文创，不适合长视频或无需确认的一键批量。","tools":[],"skillType":"skill","sortWeight":960,"downloads":1051,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/0d32c72b-1c15-4fbd-b8f6-2191d6d47f46.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"cultural-pattern-visual-system-generator","version":"1.0.4","hash":"sha256:c0c55c5d9e4b2765fee2b2707e680c556b9aa7fddc8127a3c1a2ac73437b3b97","summary":"Extracts cultural motifs into pattern masters, repeats, colorways, product mockups, packaging, wallpapers, and a short motion showcase.","summaryZh":"从文化装饰提取纹样，分步生成母版、连续纹样、配色、样机、包装、壁纸与动效。","description":"Extract cultural, architectural, or craft references into reusable patterns, colorways, mockups, packaging, and motion showcases. Use for cultural-brand visual systems, not full brand strategy.","tags":["Commercial Ad","E-Commerce"],"tagsCn":["商业广告","电商"],"creator":"","triggerWords":["文化纹样","纹样提取","花砖纹样","铁艺纹样","建筑纹样","文创品牌","品牌样机","连续纹样","配色方案","文创产品"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"文化纹样视觉系统生成器","tagEn":"Commercial Ad","tagCn":"商业广告","completeTagsEn":["Commercial Ad / Creative Generation","E-Commerce / Creative Generation"],"completeTagsCn":["商业广告 / 创作生成","电商 / 创作生成"],"descEn":"For cultural brands, designers, and creative teams that want to turn architectural, folk-craft, tile, ironwork, wall, gate, or ornamental references into a reusable visual system. The Skill processes each reference independently, producing motif masters, clean extraction masters, seamless repeats, four labeled colorways, category-matched product mockups, packaging, cultural merchandise, wallpapers, and a short motion showcase. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once when needed, then replaced by another available model. It is for pattern extraction and brand extension, not full brand strategy or long campaign production.","descCn":"面向文创品牌、设计师和创意团队，将建筑、民艺、花砖、铁艺、墙体、柱子、门窗等装饰参考转化为可复用的视觉系统。Skill 严格分步处理每张参考图，依次生成独立纹样母版、白底提取母版、连续纹样、四套带十六进制标注的配色、匹配品类的产品样机、包装与文创周边、壁纸和短动效。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。适合纹样提取、文创产品和品牌展示，不替代完整品牌策略或长篇 campaign。","tools":[],"skillType":"skill","sortWeight":970,"downloads":1030,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/1cacf030-34de-4470-ae8d-9cc3daed7e84.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"trendy-3d-ins-card","version":"1.0.2","hash":"sha256:84caf704cad95ff96b1c9c969c784104ca3f7991b78f04ad73093f00e67ef6fe","summary":"Transforms a photo into a minimalist social-frame 3D figurine card and optionally extends the accepted card into an interactive animation.","summaryZh":"把照片变成极简社交框3D手办卡，确认后可延展为固定镜头互动动画短片。","description":"Turn a portrait, pet, mascot, or character photo into a minimalist social-frame 3D card, then optionally create a fixed-camera interaction video. Use for short social visuals, not complex stories.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["social app 3D character card","photo-to-3D social UI"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"潮流3D卡片","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For creators who want a portrait, outfit, pet, mascot, or character image transformed into a minimalist social-frame 3D figurine card. The workflow reads identity, facial impression, accessories, outfit, and pose, creates a low-poly soft-vinyl figure with generous card margins and a clean backdrop, then waits for approval. If motion is requested, it offers interaction directions such as face poke, inside-frame dance, hand grab, swipe feedback, or a custom idea, and generates a fixed-camera animation while preserving the frame. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once on failure, then replaced by another available model.","descCn":"面向想把人像、穿搭、宠物、吉祥物或角色图转成社交媒体视觉卡片的创作者。流程提取主体身份、脸型印象、配饰、服装结构和姿态，生成低模软胶 3D 手办与极简社交框静态卡片，确认后再根据戳脸、框内跳舞、拉手、滑动反馈或自定义想法设计互动方向。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。动画保持外层边框和主体身份稳定，适合短视频分享，不用于复杂多镜头剧情。","tools":[],"skillType":"skill","sortWeight":980,"downloads":1035,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/afb24c03-a7fc-48ee-a728-ec0824e667b1.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"oriental-diffusion-aesthetic","version":"1.0.2","hash":"sha256:eaa29d8f1e955e16f2e123847e681f392c3a01014b0c508a1df562c8c1ec3f88","summary":"Transforms an uploaded image into an Eastern diffusion still, then optionally extends the approved still into a restrained short video.","summaryZh":"先将上传图转成东方弥散静图，确认后再延展为克制的循环短视频，保持主体色域与轮廓。","description":"Transform an uploaded image into an Eastern diffusion still, then optionally extend it into a restrained short video. Use for poster-like image transformations, not full ads or story films.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["Eastern diffusion aesthetic","diffusion aesthetic","东方弥散美学","东方雾化风格","image first video later","still then video"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"东方弥散美学","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For creators who want an uploaded product, person, object, animal, or character turned into an Eastern diffusion poster before optional motion. The workflow extracts the silhouette and visual anchors, keeps the source color gamut, simplifies the background into a dark paper-like base, and confirms the still image first. If video is requested, it uses a fixed camera, keeps the subject stable, and creates a relay-style outward color wave within the source palette. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once on failure, then replaced by another available model. Not for full ads, KOC content, or multi-shot stories.","descCn":"用于把上传的商品、人物、器物、动物或角色图先转成东方弥散静图，再在确认后延展为短视频。流程提取主体轮廓和关键锚点，保留原图已有色域，把背景压成暗底纸感并完成居中、符号化构图；静图确认后才进入视频阶段。视频默认使用 MiniMax-H3，固定机位、主体基本不动，只在原图色域内做接力式外涌循环。用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。不适合完整广告、KOC 或多镜头剧情。","tools":[],"skillType":"skill","sortWeight":990,"downloads":1027,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/492e45f1-12f4-4f1e-b37b-6b7c89a62761.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"sitcom-story-video","version":"1.0.0","hash":"sha256:7624f22594fa86cf4644208e8487fd02c26ad8c0ef276009f0273836207cf27c","summary":"Turns sitcom premises, scripts, or references into continuity-safe shorts with comic beats, shot prompts, production QA, and optional post-final instrumental BGM design.","summaryZh":"把情景喜剧素材整理成含喜剧节拍、镜头提示词、连续性检查和成片后纯器乐 BGM 设计的短片制作包。","description":"Turn a premise, script, storyboard, or references into an AI sitcom short with consistent characters, comic beats, shot prompts, and continuity QA. For narrative comedy, not generic prompts.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["AI sitcom","sitcom short","comedy short","narrative comedy short","sitcom video","comedy video","add music to final video","sound design","sitcom sound mix","canned laughter","comedy sound effects"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"情景喜剧故事视频","tagEn":"Professional Film","tagCn":"专业影视","completeTagsEn":["Professional Film / Full Production"],"completeTagsCn":["专业影视 / 成片制作"],"descEn":"For creators making narrative AI sitcom shorts from a premise, script, storyboard, or visual references. The workflow locks production choices, a durable comic engine, and a retro indoor television look; builds continuity-safe character, set, and prop bibles; prepares a timed beat sheet, cut map, shot prompts, and review assets; then guides shot generation, variant selection, tail-frame handoffs, editing, audio intelligibility, and final QA. Optional instrumental BGM is designed only after the first final cut. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once for the diagnosed failure, then replaced by another available model. Best for short situation comedies, not standalone images, generic prompts, or long-form drama.","descCn":"面向制作叙事型 AI 情景喜剧短片的创作者，可输入梗概、剧本、分镜或视觉参考。流程先锁定制作规格、喜剧引擎和复古室内电视风格，建立角色、场景与道具连续性锚点，再编写时间化节拍表、剪辑图和逐镜提示词，审核中间资产后逐镜生成、筛选版本、衔接尾帧并完成剪辑与声音验收。需要时可在成片后加入带时间码的纯器乐 BGM。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换可用模型。适合情景喜剧短片，不用于单图、泛化提示词或长篇剧情制作。","tools":[],"skillType":"skill","downloads":1292,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/90f805fe-bbed-4edb-8b3d-005da35c18c1.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"90s-cel-animation-city-pop-short","version":"1.0.0","hash":"sha256:5bb69031ec8c543c439ab58bce6f1f214690a116993a7b694b387881f27d3cc1","summary":"Turns stories and visual references into connected 1990s cel-animation City Pop shorts with locked style, continuity assets, segment storyboards, music, sound design, and final editing.","summaryZh":"将剧情与视觉参考转化为统一风格、连续资产、分段分镜、音乐拟音和剪辑完整的复古都市赛璐璐动画短片。","description":"Create 30–90s, 16:9 Japanese 1990s cel-animation City Pop shorts from stories and references, with continuity assets, segment storyboards, video, sound, and editing; not for stills or simple i2v.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["90s cel animation","City Pop animation","retro urban anime short","cel animation short","90年代赛璐璐动画","复古都市动画","City Pop动画短片"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"复古都市赛璐璐短片","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For creators making 30–90 second horizontal urban animation shorts inspired by Japanese 1990s flat graphic cel animation and City Pop. Users provide a story, character or scene images, reference videos, and optional audio. The workflow analyzes source roles, locks a Style Master and timeline, builds reusable characters, props, and environments, prepares one 2×2 storyboard per narrative segment, generates approved video segments, then adds music, urban sound, transitions, and a unified edit. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once for the diagnosed failure, then replaced by another available model. Best for travel impressions, city memories, landmarks, transit, and urban montages; not still images, simple i2v, subtitle-only work, full music videos, game PVs, or projects over 90 seconds.","descCn":"面向制作30–90秒、16:9横屏复古都市动画的创作者，把剧情、人物或场景图片、参考视频和可选音频转化为日本90年代平面图形化赛璐璐 City Pop短片。流程先分析素材作用并锁定 Style Master 与时间轴，再建立人物、道具和场景资产，为每个剧情段制作并确认一张2×2分镜，随后逐段生成视频，加入音乐、城市拟音、转场并统一剪辑。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换可用模型。适合旅行印象、都市记忆、地标与交通蒙太奇，不用于静图、简单图生视频、完整长 MV、游戏 PV 或超过90秒项目。","tools":[],"skillType":"skill","downloads":1143,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/099c7370-3287-42d3-b463-d443e2442fea.gif","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"stamp-photo-motion","version":"1.0.2","hash":"sha256:ab8a935fcc4007170be389466ffc4ea7dcfee9d9a26a855566de7a6e84445679","summary":"Restyles a photo, builds a split postage-stamp composite, and animates the scene with a fixed camera while preserving the frame and style boundary.","summaryZh":"把照片重绘并合成邮票双风格画面，再以固定镜头生成保持边界的动画。","description":"Restyle a photo, build a split postage-stamp composition, and animate it with a fixed camera while preserving layout and the style boundary. Use for mixed-media collage scenes.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["postage stamp","stamp frame","dual style","picture-in-picture","Korean naive illustration","flat illustration motion","邮票框","双风格","画中画","韩式稚拙插画"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"邮票照片动效","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For creators who want a daily photo turned into a mixed-media postage-stamp animation. The workflow first restyles the source as a naive flat illustration, then composites a centered stamp frame whose interior uses the illustration and exterior keeps the original photo, aligning shared buildings, horizons, or routes. After approval, it animates the scene with a fixed camera while the stamp and style split remain stable. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once on failure, then replaced by another available model. Best for collage-like scenes, not complex acting, multi-shot stories, or pure retouching.","descCn":"面向想把日常照片变成有趣分享素材的创作者。流程先将原图重绘为韩式稚拙扁平插画，再以原图为底叠加居中邮票边框，使框内呈插画、框外保留照片，并让共同建筑、地平线或路径在内外准确对齐。静态合成确认后，使用固定镜头让场景自然运动，同时保持邮票框位置和内外画风边界稳定。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败针对原因重试一次后切换其他可用模型。不适合复杂角色表演、多镜头剧情或纯修图。","tools":[],"skillType":"skill","sortWeight":1000,"downloads":1031,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/b8345690-5e42-43b6-8fcb-2f71c1a27ef6.mp4","authorEn":"Kido G","authorCn":"Kido G","source":"community","installed":false,"updateAvailable":false},{"name":"half-narrated-2d-short-drama","version":"1.0.0","hash":"sha256:af590eeef54ef31b3013f46719910a06a0a928e84229ebbe02e3bcca9595e342","summary":"Locks narration and dialogue windows during storyboard design, generates consistent narration with original dialogue assets, and hands the validated package to the user for editing.","summaryZh":"在分镜阶段锁定旁白与对白窗口，生成统一旁白和原声对白素材，交付用户自行剪辑。","description":"Create a 2D cel/webtoon half-narrated drama: lock speech windows, preserve original dialogue, hand off narration, and let the user edit. MiniMax-H3 is default; check explicit alternatives first.","tags":["Short Drama","Animation"],"tagsCn":["短剧漫剧","动画"],"creator":"","triggerWords":["2D half-narrated short drama","cel animation drama","webtoon short drama","animated narrated drama","2D动画半解说短剧","2D半解说短剧","国漫webtoon"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"2D动画半解说短剧","tagEn":"Short Drama","tagCn":"短剧漫剧","completeTagsEn":["Short Drama / Planning","Short Drama / Creative Generation","Animation / Creative Generation"],"completeTagsCn":["短剧漫剧 / 计划制定","短剧漫剧 / 创作生成","动画 / 创作生成"],"descEn":"For creators making a 2D cel or webtoon half-narrated short drama from a story, script, reference video, or existing assets. The workflow locks plot, assets, resolution, narrator voice, and speech windows before generation, binds every shot to approved characters and scenes, validates original dialogue audio, and delivers a narration master, dialogue tracks, timing metadata, and editing handoff notes. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once for the diagnosed cause, then replaced by another available model. Best for staged short-drama production; it does not perform final cutting, mixing, subtitle burn-in, or export QC.","descCn":"面向需要制作2D赛璐珞或国漫 webtoon 半解说短剧的创作者。用户提供故事、剧本、参考视频或已有素材，流程先锁定剧情、资产、清晰度和旁白音色，规划每镜旁白与对白窗口，再绑定角色场景资产并逐镜生成视频，验证原声后交付旁白母轨、对白轨和时间元数据，由用户自行剪辑。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换可用模型。适合分阶段制作，不负责最终剪辑、混音、字幕烧录或导出质检。","tools":[],"skillType":"skill","sortWeight":1010,"downloads":2072,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/d58b3975-d059-4c5e-90e7-3d6431761ab7.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"wool-felt-story-short","version":"1.0.1","hash":"sha256:8700cc1c39bfae05c66946a0c2f3727d3e3ea621cb63c675c3264228ec2095b3","summary":"Turns story copy into a gated wool-felt short with locked format and image/video resolution choices, non-overlapping narration and dialogue timing, reusable visual assets, validated shot audio, and separate final cuts.","summaryZh":"将故事文案转为分阶段确认的羊毛毡短片，锁定画幅、图片与视频清晰度、旁白音色和非重叠音频时间轴，并交付独立成片。","description":"Create flexible wool-felt story shorts from narration or scripts with staged review, resolution choices, timed voices, reusable assets, shot generation, and assembly. Not for stills or miniatures.","tags":["Animation","Sound & Music"],"tagsCn":["动画","音频音乐"],"creator":"","triggerWords":["wool felt animation","felt story short","narrated felt video","wool-felt short","handmade felt story","realistic-scale felt world"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"羊毛毡故事短片","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation","Sound & Music / Creative Generation"],"completeTagsCn":["动画 / 创作生成","音频音乐 / 创作生成"],"descEn":"For creators turning narration, scripts, or story copy into portrait or landscape wool-felt story shorts. Users may add character, scene, prop, and style references. The workflow locks an approved aspect ratio, asks for 1K or 2K image assets and 768P or 2K video, builds reusable felt elements, plans narration and dialogue without unintended overlap, reviews shots one at a time, preserves natural sound effects, and assembles a no-BGM cut before offering a separate BGM version. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once for the diagnosed failure, then replaced by another available model. Best for realistic-scale felt worlds and continuity-sensitive stories, not standalone images, one-off clips, realistic commercials, miniature dioramas, puppet stages, or toy displays.","descCn":"面向把旁白、脚本或故事文案制作成竖版或横版羊毛毡短片的创作者，可补充角色、场景、道具和风格参考。流程先锁定画幅，并在相应生成节点确认图片与视频清晰度；随后建立可复用羊毛毡元素，规划纯旁白、纯对白或混合模式的非重叠音频时间轴，逐镜生成并确认画面、原声对白和环境音，先合成无 BGM 成片，再按需输出独立配乐版。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换可用模型。适合现实比例羊毛毡叙事，不用于单图、写实广告或微缩玩偶舞台。","tools":[],"skillType":"skill","sortWeight":1020,"downloads":1827,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/a3af6f27-9d65-49b7-b278-73cd7fd5b1ef.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"half-narrated-3d-short-drama","version":"1.0.0","hash":"sha256:cccc82f137584ed542a071be8e70a56f7e8b9bb84b768b11f8c350efd1a671f7","summary":"Turn a story or script into a guided 3D manhua short-drama package with dialogue video and separate narration audio.","summaryZh":"把故事或剧本整理成含对白视频和独立旁白音频的可剪辑3D漫剧短剧制作包。","description":"Create a guided 3D manhua half-narrated short drama from a story or script. Plan narration and dialogue, generate anchored shots, and deliver a dialogue video with separate narration audio.","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":["3D manhua drama","half-narrated 3D drama","3D short drama","manhua short film","narrated 3D drama","fantasy manhua drama"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"3D漫剧半解说","tagEn":"Short Drama","tagCn":"短剧漫剧","completeTagsEn":["Short Drama / Planning","Short Drama / Creative Generation","Short Drama / Post-production"],"completeTagsCn":["短剧漫剧 / 计划制定","短剧漫剧 / 创作生成","短剧漫剧 / 后期处理"],"descEn":"For creators developing a 3D manhua short drama from a premise, script fragment, reference image, or existing assets. Confirm the story and delivery choices, reconcile every source beat, reserve narration and dialogue windows, build reusable character and scene anchors, generate measured audio and image-to-video shots, then assemble a dialogue-only video in the approved order. Deliver narration, its transcript, and placement timing as a separate package for user-led editing. The workflow prioritizes continuity and clear review gates. Best for short episodic 3D manhua; not live action, pure 2D animation, narrated master mixing, or long-form series planning.","descCn":"面向需要制作 3D 漫剧短剧的创作者，可从故事、剧本片段、参考图或已有素材开始。Skill 先确认剧情与交付选项，完成剧本对账，在分镜阶段预留旁白和对白时间窗，建立角色与场景锚点，生成定时音频和图生视频镜头，再按确认顺序拼接无旁白对白视频。旁白作为独立音频包交付，方便用户自行剪辑。适合短篇连续 3D 漫剧；不负责真人实拍、纯二维动画、旁白混音母版或长篇系列统筹。","tools":[],"skillType":"skill","sortWeight":1030,"downloads":1766,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/3c465f19-ad82-4ee4-bc80-7532f52df93c.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"half-narrated-live-action-short-drama","version":"1.0.0","hash":"sha256:2908431c47282bcd8c7e26a13646a879805da0b137e4a104f0869442b220830e","summary":"Turns stories or scripts into narrated live-action short-drama episodes with character dialogue, continuity-controlled shots, user-confirmed video resolution, and optional subtitles.","summaryZh":"将故事或剧本制作成由旁白推进、角色对白补充的真人短剧单集，并确认视频分辨率后完成连续性制作。","description":"\"Create narration-led live-action drama from scripts and references: continuity assets, timed shots, assembly, narration, and optional subtitles. MiniMax-H3 default; check explicit alternatives.\"","tags":["Short Drama"],"tagsCn":["短剧漫剧"],"creator":"","triggerWords":["半解说真人短剧","第一人称旁白短剧","第三人称旁白短剧","真人写实短剧","竖屏短剧","重生短剧","末世短剧","悬疑短剧","narrated live-action short drama"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"半解说真人短剧","tagEn":"Short Drama","tagCn":"短剧漫剧","completeTagsEn":["Short Drama / Creative Generation"],"completeTagsCn":["短剧漫剧 / 创作生成"],"descEn":"For creators producing narrated live-action short dramas where voiceover drives the story and character dialogue carries the conflict. Provide a premise or script with optional character, scene, and audio references. The workflow locks visual direction, builds continuity assets, plans timed narration and dialogue shots, generates keyframes and videos, and aligns a standalone narration track to the assembled timeline. MiniMax-H3 is the default video model; if the user explicitly selects another model, check its capabilities first, retry a failed shot once for the diagnosed cause, then switch to another available model. It delivers a short-drama video, narration track, and optional subtitles, not animation, pure commentary, or ordinary editing.","descCn":"面向制作半解说真人短剧的创作者，适用于由旁白推进剧情、角色对白补充冲突的内容。用户提供故事或剧本，也可补充角色、场景和音频参考。流程先锁定视觉方向，建立连续性资产，规划带时间码的旁白与对白镜头，生成关键帧和视频，并将独立旁白对齐到最终时间轴。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换可用模型。交付短剧视频、旁白音轨和可选字幕，不用于动画、纯解说或普通剪辑。","tools":[],"skillType":"skill","sortWeight":1040,"downloads":1608,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/b5efbc58-d361-49b1-9381-2b20863831a4.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"future-sci-fi-cinematic-generator","version":"1.0.0","hash":"sha256:a65cd708655299d3704d313bdaf79206568022a4d3e1f6ab48d4128e1c52e7ec","summary":"Turns a confirmed future-sci-fi brief into continuity-safe characters, scenes, storyboards, timed video segments, and an assembled realistic short film.","summaryZh":"将确认的未来科幻故事转为连续的人物场景资产、分镜、分段视频和写实短片成片。","description":"Create a realistic future-sci-fi short from a brief and references: continuity-safe assets, storyboards, segmented video, and final assembly. MiniMax-H3 is default; check explicit alternatives first.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["future sci-fi cinematic generator","future sci-fi film","realistic sci-fi short","live-action sci-fi","未来科幻质感影视生成器","未来科技感短片","未来科幻写实短片","超写实科幻电影"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"未来科幻质感影视生成器","tagEn":"Professional Film","tagCn":"专业影视","completeTagsEn":["Professional Film / Creative Generation"],"completeTagsCn":["专业影视 / 创作生成"],"descEn":"For creators who need a realistic live-action future science-fiction short from a story brief. Provide a synopsis, optional character or scene references, duration, paragraph count, aspect ratio, language, and style preference. The workflow locks the visual direction, audits references, builds complete character, species, prop, and scene assets, binds them to a continuity storyboard, generates timed video segments, and assembles the approved film. MiniMax-H3 is the default video model; if the user explicitly selects another model, check its capabilities first, retry a failed segment once for the diagnosed cause, then switch to another available model. Best for one 16:9 narrative short; not for anime or game PVs, cyberpunk-first pieces, music videos, title sequences, or simple image-to-video motion.","descCn":"面向需要制作超写实真人未来科幻短片的创作者。用户提供故事梗概、人物或场景参考图、时长、段落数、画幅和风格偏好。流程先锁定视觉方向与规格，再建立人物、物种、道具和场景资产，制作连续性分镜，分段生成并按时间轴合成成片。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换可用模型。适合16:9叙事科幻短片，不适合二次元、赛博朋克主导、音乐视频、片头标题序列或简单图生视频。","tools":[],"skillType":"skill","sortWeight":1050,"downloads":1294,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/47f97192-a363-4eba-be9a-331062e6450d.mp4","authorEn":"MiniMax Design","authorCn":"MiniMax Design","source":"official-featured","installed":false,"updateAvailable":false},{"name":"digital-drawing-timelapse","version":"1.0.0","hash":"sha256:78fa915ae12511d7aded43aacf22f53968c0021b2d0ebdfd2b1450021ee923b2","summary":"Turns one finished artwork and one prompt into a continuous 15-second digital drawing-process video for illustration showcases.","summaryZh":"根据一张成品图和提示词，生成15秒连续数字绘画过程视频，适合插画展示。","description":"Generate a 15-second digital drawing timelapse from one prompt and one finished-artwork image. Use MiniMax-H3 by default; check explicitly selected alternatives first.","tags":["Animation"],"tagsCn":["动画"],"creator":"","triggerWords":["digital drawing timelapse","speedpaint video","drawing process video","手绘过程","手绘延时摄影","绘画过程视频","数字绘画过程"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"数字绘画延时摄影生成器","tagEn":"Animation","tagCn":"动画","completeTagsEn":["Animation / Creative Generation"],"completeTagsCn":["动画 / 创作生成"],"descEn":"For illustrators and creators who want a finished digital artwork presented as a drawing-process timelapse. Provide exactly one finished-artwork image and one prompt. The workflow verifies both inputs, preserves the prompt without adding creative constraints, confirms duration, aspect ratio, and resolution, then generates one continuous 15-second speedpaint-style video and checks that the drawing visibly progresses from blank canvas to the finished reference. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once for the diagnosed cause, then replaced by another available model. Best for digital-art showcases, not interface tutorials, long courses, multi-asset work, or live-action photo conversion.","descCn":"面向插画师和内容创作者，将一张数字绘画成品图转成15秒绘画过程延时视频。用户提供成品图和一段提示词，流程确认输入后保留原提示词，锁定画幅、时长与清晰度，直接生成连续速涂过程并复查画面连贯性。默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换可用模型。适合数字绘画展示，不适合界面教程、长课程、多素材制作或真人照片转手绘。","tools":[],"skillType":"skill","sortWeight":1060,"downloads":1041,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/4bdf5b99-2395-4063-8e3b-070d9e38ac66.gif","authorEn":"aimikoda","authorCn":"aimikoda","source":"community","installed":false,"updateAvailable":false},{"name":"windows-xp-dialog-story","version":"1.0.0","hash":"sha256:fc5bb52840cbac2c75af6ca686df6a6ad46c8997b2695b07b8462ec25e23d093","summary":"Turn retro Windows XP dialog states and character actions into readable, escalating visual shorts with a clear resolution.","summaryZh":"将复古 Windows XP 弹窗状态与角色动作编排成可读、递进且有结算的视觉叙事短剧。","description":"Create Windows XP Luna dialogs or popup-driven action shorts for retro UI, popup storytelling, error escalation, progress interactions, and terminal results; not for full desktop recreations.","tags":["Professional Film"],"tagsCn":["专业影视"],"creator":"","triggerWords":["Windows XP dialog","XP popup video","popup animation","system dialog video","retro system UI video","Windows XP 弹窗视频","弹窗动画","系统弹窗玩法","弹窗叙事"],"guidePrompt":"为我解释一下这个技能的最佳使用方式。","guidePromptEn":"Show me the best way to use this skill with a few examples.","displayNameZh":"Windows XP 弹窗视觉短剧","tagEn":"Professional Film","tagCn":"专业影视","completeTagsEn":["Professional Film / Full Production"],"completeTagsCn":["专业影视 / 成片制作"],"descEn":"For creators who want to turn retro Windows XP Luna dialogs into visual narratives. Starting from a brief concept, dialog artwork, character references, or a script, the workflow builds error, warning, progress, interruption, and settlement states, then maps character actions to popup arrivals, stacking, collisions, progress changes, and system transitions on a readable timeline. It can deliver a standalone dialog image or an early-DV-style action short while preserving supplied copy, popup legibility, spatial occlusion, causal state changes, and camera rhythm. MiniMax-H3 is the default video model; an explicitly selected alternative is capability-checked, retried once for the failure cause, then replaced by another available model. It is not for functional interfaces, full desktop recreations, or exact reference copying.","descCn":"面向希望把复古 Windows XP Luna 弹窗转化为视觉叙事的创作者，可从简短创意、弹窗素材、角色参考或剧本出发，先建立报错、警告、进度、中断与结算状态，再把人物动作与弹窗出现、层叠、碰撞、进度变化和系统转场映射到清晰时间线，输出独立静态弹窗或具有早期 DV 质感的动作短片。流程强调文案准确、弹窗可读、空间遮挡、状态因果和镜头节奏。视频默认使用 MiniMax-H3；用户明确指定其他模型时先检查能力，失败按原因重试一次后切换其他可用模型。不用于可运行界面、完整桌面复刻或照搬参考视频。","tools":[],"skillType":"skill","sortWeight":1070,"downloads":1015,"coverUrl":"https://cdn.hailuoai.com/hailuo-video-web/public_assets/47648246-ff13-4af7-bd01-edd2894294c6.mp4","authorEn":"山音","authorCn":"山音","source":"community","installed":false,"updateAvailable":false}],"total":109}