用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/LYL1015/JarvisHub --skill flova-miniature-world-short命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
Use when the user asks JarvisHub AI Chat or a `webHero` workflow to generate a website, landing page, homepage, brand-style web design, preview-first webpage workflow, or wants section screenshots, webpage assets, and final HTML to be produced as one staged flow.
把 assets/demo 提炼成运行时可用的视觉连续性方法论:先锁资产/角色/场景/镜头语义,再做扩镜与视频。(素材生成场景:角色一致性、视觉锚点锁定、单素材生成前的锚点规范与一次只改一个变量。)
基于用户提供的设计稿图片还原真实网页。该 skill 可用于任意前端项目,但强依赖本地已启动且已授权的 JarvisHub vision 工作流;视觉模型由服务端默认配置决定,支持外部传入 vision prompt,且不可静默降级。
基于 SOC 职业分类
正在显示 SKILL.md
| name | flova-miniature-world-short |
| description | Use when 用户要制作微缩世界创意短片、移轴摄影美学、微型模型场景、温柔情感故事、节日祝福视频、60-90 秒竖屏微缩生活故事。 |
用于微缩世界创意短片:以微型道具、模型场景和移轴摄影美学讲一个有情绪弧光的温柔小故事,适合节日祝福、生活故事、暖调情感短片。
imageUrl / videoUrl,不能把提交态当完成态。critic sub-agent:只读取真实媒体并评审,不生成、不补素材。blocked 项,除非本轮工具列表明确暴露对应能力。Use this skill for:
Do not use it for real documentary, regular drama, or wool felt full-size animation.
Follow the JarvisHub Execution Model. Do not assume a fixed model menu, membership state, or exact duration support. If exact ratios/durations are unsupported, disclose and provide closest runnable prompts.
Confirm first:
Do not plan the story before these production specs are known.
Pause after specs, storyboard, elements, completed clips/audio plan or assets, and rough cut.
Build 4-6 story beats:
Each beat should define:
Typical 60s film: 10-12 shots. Typical 90s film: 14-18 shots.
All images and video should feel like miniature model photography:
Characters:
Scenes:
Props:
Each shot should include:
Default shot duration: 4-8s. Emotional high point may hold up to 10s.
Camera:
Avoid fast cuts and aggressive handheld motion.
Image prompt should include:
miniature model photography, tilt-shift lens look, shallow depth of field, handmade miniature set, [characters/action/scene], delicate material texture, soft diffused lighting, warm muted palette, no full-size real-world scale confusion, no random text, no watermark
Video prompt should include:
Use the miniature character and scene references. [shot action and emotion]. Gentle [camera move], tilt-shift miniature depth of field, soft diffused light, delicate model texture, warm quiet pacing, no subtitles, no random text, no watermark, no music unless audio is intended in the clip.
Maintain scale logic:
BGM:
Narration optional:
If video clips include built-in audio and separate narration exists, avoid overlapping unwanted speech/noise.
Check:
If it reads as full-size live action, regenerate with stronger miniature model and tilt-shift constraints.
Return:
spec,story_beats,key_elements,shot_list,prompts,audio_plan,media,assembly_plan,review.Do not present element images as finished video.