人物微表情表演
SkillMediaOrganizes character images, script excerpts, or emotion prompts into time-segmented micro-expression performance guidance, and generates selectable performance clips after user confirmation. Suited for nuanced character acting; not for full screenwriting, facial rigging, lip-sync editing, or final c
Available today. Use it from your connected AI after setup.
No other account needed.
Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
Then ask your AI: use the 人物微表情表演 skill
What this skill tells your AI
The instructions your AI receives, as published by qxryz/workflowgenerator in skills/library/micro-expression-video-generator/SKILL.md and read by ahel’s review.
当用户想让角色的情绪表演更自然、更细腻、更像真实演员时,使用这个 Skill。
输入方式
- 图片模式:给一张角色图,继续设计表演。
- 剧本模式:给一段剧本、分镜或 prompt,只增强表演层。
- 情绪短语模式:给一句情绪或关系状态。
参考资料库
细节规则都放在这些文件里:
references/source-notes.mdreferences/performance-prototype-library.mdreferences/emotion-route-library.mdreferences/muscle-dispatch-library.mdreferences/video-prompt-guardrails.mdreferences/tempo-density-guide.mdreferences/climax-reset-patterns.md
流程
- 先读用户输入,保留已有角色、场景、台词和运镜。
- 如果缺少镜头时间,就先问镜头时间;如果缺少表演强度,也先问。
- 写出简洁提示词包并交给用户审阅;只有确认后才继续生成可选表演片段。优先使用 MiniMax H3,因为它能满足本流程所需的图片、音频参考和表演控制能力。用户指定其他模型时,只要满足已确认要求就遵循其选择;如果能力不符,说明差异并让用户选择等效方案。
边界
这个 Skill 只负责表演提示词和情绪表演方向,不替代完整编剧、面部绑定、口型编辑或最终合成。
Signals
- GitHub stars
- 26
- Last commit
- Sep 2026
Advanced
- Item type
- skill
- Key
micro-expression-video-generator- Source
- github.com/qxryz/workflowgenerator