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-east-asian-arthouse-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。
正在显示 SKILL.md
| name | flova-east-asian-arthouse-short |
| description | Use when 用户要创作东亚文艺片风格短片、情绪氛围驱动视频、克制表演、长镜头、留白、声画分离、物件隐喻或内收情感表达。 |
用于东亚文艺片式短片:以情绪轨迹而非剧情事件为核心,设计克制表演、长镜头、留白、物件转喻和声画分离。
The goal is not "what happened" but "what is slowly changing inside someone."
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 fast plot-driven dramas, product ads, documentaries, or social montage.
Follow the JarvisHub Execution Model for assets, video clips, review, and assembly. Treat audio as plan or bound asset unless a real audio tool is exposed.
If the current video model struggles with intense action or facial emotion, use the replacement strategies in this skill instead of forcing direct generation.
Pause after:
Before plot, define:
Use a breathing structure:
If a proposed scene is only plot information, convert it into an emotion beat or remove it.
Characters:
Scenes:
Objects:
Provided references must be preserved.
Allowed camera:
Recommended duration:
Avoid:
Every shot must include:
Use internalized expression:
Avoid:
If strong emotion is necessary, use:
Use:
References may include broad qualities of Kore-eda, Hou Hsiao-hsien, Lee Chang-dong, Wong Kar-wai mood color, without copying exact scenes.
Avoid:
Image/keyframe prompt:
[character or object micro action] in [lived-in East Asian space], [time/light source], [shot scale], [composition and negative space], [soft color/film texture], restrained emotion, no melodrama, no subtitles, no random text, no watermark
Video prompt:
[allowed camera movement]. [subject performs a small continuous action]. [environment/light changes slightly]. [emotional phase]. slow pacing, breathing space, no fast cuts, no subtitles, no random text, no watermark, no music unless audio is intentionally generated inside clip
Do not describe plot summaries in video prompts. Describe visible behavior.
Audio is emotional continuity:
Use sound-image separation:
Editing:
Check:
If a shot feels too explanatory, replace it with object, silence, or aftermath.
Return:
emotional_trajectory,final_spec,key_elements,shot_list,keyframe_prompts,video_prompts,audio_plan,media,assembly_plan,review.Do not present a plot outline as a finished arthouse film.
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