소스 정보
- 저장소
- LYL1015/JarvisHub
- 최근 소스 활동
- 2026년 7월 26일 16:09
- 감지된 SKILL.md 언어
- 영어
- 스타
- 462
- 포크
- 42
설치 방법
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
소스 파일 검토
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
메뉴
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/LYL1015/JarvisHub --skill flova-pet-anthropomorphic-vlog명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
SKILL.md 표시 중
| name | flova-pet-anthropomorphic-vlog |
| description | Use when 用户上传猫狗等宠物照片,要制作萌宠拟人打工 Vlog、猫猫情感短片、宠物职场梗、双宠物日常或系列化宠物剧情。 |
用于萌宠拟人 Vlog:用真实宠物身份作为视觉锚点,把人类处境、职场梗或集体情绪翻译成宠物能自然表演的短视频。
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 normal pet portrait generation or human POV romance scenes.
Follow the JarvisHub Execution Model for image analysis, image generation, video generation, review, and assembly. Treat audio as plan or bound asset unless a real audio tool is exposed.
Require at least one clear pet reference image unless the user explicitly wants a fully fictional pet. For identity-preserving work, the image must show face, body/coat, and distinguishing marks clearly enough.
If multiple pet images are provided, classify:
Ask for confirmation before binding references.
The pet may be anthropomorphized, but must remain an animal character.
Avoid:
Workplace satire should be visual, light, and fictional.
| Mode | Use For | Core Rule |
|---|---|---|
pet_workday_vlog | job/persona-based funny daily video | profession DNA + pet visual DNA |
pet_emotional_short | nostalgia, reunion, childhood, healing | cat body + human situation + collective emotion |
multi_pet_vlog | two pets, cat/dog pair, group interaction | each pet has separate identity card |
Pause after visual DNA/spec, story direction, storyboard, character reference, first clip/batch, and final assembly.
Extract:
This becomes the identity authority. Do not rewrite pet appearance per shot.
Confirm one:
light: mostly natural pet behavior, small props/costume.medium: pet wears work accessories and interacts with scaled props.strong: pet performs stylized human-like scenes, but still visibly animal.Even in strong mode, preserve pet anatomy, coat, and recognizable identity.
Define profession DNA:
Useful profession routes:
Every joke must be visible. For example, "requirements changed" needs a screen/message/sticky-note reaction, not just a subtitle.
Use the formula:
pet body x human situation x collective emotion
Choose:
Three layers:
Avoid heavy narrated sentiment. Let action, silence, and a small final gesture carry the emotion.
Each shot must include:
shot_id,scene_location,pet_character_ids,visible_pet_behavior,anthropomorphic_action_if_any,scaled_props,camera_style,duration,audio_notes,continuity.Default:
Use natural animal behavior first: walking, sitting, pawing, sniffing, blinking, tail movement, hiding, stretching, napping. Add human-like props only when needed.
For each pet:
Scene assets:
Default:
For workday comedy:
For emotional shorts:
Use reference placeholders for pet identity. Do not re-describe pet appearance in every shot unless needed for identity correction.
Video prompt order:
Always include:
preserve the pet identity from the reference image, realistic animal texture, pet-scale props, no subtitles, no random text
For human presence:
Check:
Fix one shot or reference at a time.
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。
SOC 직업 분류 기준