| name | Zhao-LibTV-Prompt |
| description | 专业分镜脚本生成器。将剧本转换为详细的分镜脚本，包含景别（特写/近景/中景/全景/远景）、机位、运镜、画面描述、音效、每句台词的声音要求（情绪强度/语速/音色/状态/停顿）、光影氛围、正反向提示词。适用于古装短剧、现代剧、影视制作等所有类型的视频分镜规划。 |
Zhao-LibTV-Prompt Skill
Overview
Professional storyboard script generator for AI video generation. Converts scripts into detailed shot-by-shot breakdowns with camera angles, movements, lighting, and voice direction.
When to Use
- User provides a script/dialogue and asks for storyboard/shots
- User wants to convert text into visual sequences
- User needs detailed camera directions for AI video generation
- User wants to plan scenes with precise timing and emotional beats
Core Workflow
Step 1: Script Analysis
Analyze the provided script for:
- Scene Structure: Identify distinct scenes and their locations
- Character Arcs: Track emotional progression throughout
- Key Moments: Pinpoint climactic/emotional peaks
- Pacing: Determine rhythm and tempo of each section
- Visual Elements: Note props, settings, costumes, weather
Step 2: Scene Breakdown
Divide script into timed segments (typically 5-15 seconds each):
- 0-X seconds: Opening/setup
- X-Y seconds: Conflict/rising action
- Y-Z seconds: Climax/resolution
- Z-END: Closing/transition
Each segment needs:
- Time range
- Scene location
- Characters present
- Emotional tone
- Key actions/dialogue
Step 3: Shot Design
For each segment, define:
Visual Composition
【景别】Shot Size (特写/近景/中景/全景/远景) 【机位】Camera Position (正面/侧面/俯视/仰视/跟随) 【运镜】Camera Movement (推/拉/摇/移/跟/甩/旋转)
Action Description
Write in shot-by-shot format:
`
Use progressive shot sizes to build tension or release it.
Step 4: Voice Direction
For EVERY line of dialogue, add:
`
【[角色]声音要求】
- 情绪强度：X% (0-100)
- 语速：描述 (快/慢/变化)
- 音色：描述 (年龄/性别/质感)
- 状态：心理/生理状态
- 停顿：精确到秒的停顿点
`
Step 5: Technical Specs
Add these sections:
`
【光影氛围强化词】
Lighting, color palette, atmosphere descriptors
【通用固定提示词】
Positive prompts for AI generation
【负向提示词】
Negative prompts to avoid
【旁白反向提示词】
Negative prompts for narration
`
Output Format
Follow this exact structure:
`markdown
[Episode Title] —— 分镜脚本
【整体基调】
- 色调：[color palette]
- 影调：[lighting style]
- 节奏：[pacing description]
- 总时长：[total seconds]
[Time Range] ｜ [Segment Title]
【景别】 [Shot size progression]
【机位】 [Camera positions]
【运镜】 [Movement types]
【画面】
- [Shot 1 description]
- [Shot 2 description]
- [Shot 3 description]
【音效】
【[角色]】
"[Dialogue]"
【[角色]声音要求】
- 情绪强度：[X]%
- 语速：[description]
- 音色：[description]
- 状态：[description]
- 停顿：
Best Practices
Shot Progression
- Build tension: 远景 → 全景 → 中景 → 近景 → 特写
- Release tension: 特写 → 近景 → 中景 → 全景 → 远景
- Quick cuts for action/comedy
- Slow pushes for drama/emotion
Timing
- Opening shots: 3-8 seconds (establish scene)
- Dialogue exchanges: 5-15 seconds per line
- Climactic moments: 10-20 seconds with multiple shots
- Transitions: 2-5 seconds
Emotional Arcs
Map emotion intensity (0-100%) across the episode:
- Beginning: Low-Medium (20-40%)
- Rising action: Medium-High (40-70%)
- Climax: Very High (70-90%)
- Resolution: Medium-Low (30-50%)
Voice Direction Tips
- Pause duration creates emphasis
- Speed changes indicate emotion shifts
- Tone quality reveals character depth
- State descriptions guide performance authenticity
Example Patterns
Pattern 1: Dramatic Reveal
特写眼睛 → 拉远至中景 → 全景展示环境
Pattern 2: Action Sequence
快速切换：近景动作 → 特写面部 → 中景全景 → 特写结果
Pattern 3: Emotional Beat
静止长镜头 → 缓慢推近 → 特写微表情 → 拉远留白
Customization
Adjust based on genre:
- Historical: Focus on architecture, costumes, traditional lighting
- Modern: Contemporary settings, natural lighting, realistic dialogue
- Fantasy: Enhanced lighting, atmospheric effects, stylized compositions
- Comedy: Faster cuts, wider shots, exaggerated expressions
Quality Checklist
Before delivering:
Advanced Techniques
Cross-Fading Emotions
Use shot progression to mirror emotional states:
- Confusion: Disjointed cuts, multiple angles
- Clarity: Single sustained shot, slow movement
- Tension: Progressive zooms, shallow depth of field
- Resolution: Wide shots, stable camera
Parallel Storytelling
For multiple characters:
- Alternate between POVs
- Use matching shots (same composition, different subjects)
- Cross-cut during climaxes
- Converge visually at resolution
Symbolic Framing
Embed meaning through composition:
- Windows/doors: Transition/opportunity
- Mirrors: Self-reflection/duality
- Heights: Power/status
- Shadows: Mystery/threat
- Light sources: Hope/clarity
Integration with LibTV
This skill works best when:
- Script is provided first
- Character profiles are available
- Visual style references exist
- Target platform specifications are known
For optimal results, combine with:
- Storyboard visualization tools
- Voice synthesis for audio previews
- Shot list management software
- Production scheduling apps
Version History
- v1.0.0 (2026-07-09): Initial release
- Complete storyboard generation framework
- Voice direction methodology
- Shot progression patterns
- Quality checklist
- Advanced techniques
Generated by Zhao-LibTV-Prompt Skill
For questions or contributions: https://github.com/Zhao-LibTV/Zhao-LibTV-Prompt
