holographic-explainer

Files (6)

• SKILL.cn.md
• SKILL.md
• meta.yaml
• hologram-library.md · references/
• narration-budget-guide.md · references/
• ref2va-format-guide.md · references/

SKILL.md (first lines)

# Holographic Explainer (全息讲解)

Turn any user-provided script and reference images into polished multi-segment H3 video prompts featuring a character presenter, holographic UI interaction, and complete narration coverage. Each segment is an independently usable Ref2VA prompt. This Skill is fully self-contained and does not depend on any other skill.

## When to Use

- A character presenter explains a topic with holographic UI interaction
- A product tutorial, tech guide, or educational explainer needs polished short-video prompts
- A narration script needs to be split into timed segments with full coverage
- Hardware/software comparisons need holographic visualization

Do not use when:
- No character reference image is available
- The user needs non-narrated B-roll without a presenter
- The user needs only a written script without video prompts

## Required Input

The user must provide:
1. **Character reference image** (Picture 1): defines presenter identity, face, hair, glasses, wardrobe, accessories.
2. **Environment reference image** (Picture 2): defines room layout, color palette, furniture, lighting mood.
3. **Narration script**: the complete text to be spoken across all segments.

Optional input:
- **Presenter name**: the on-camera identity. Defaults to "Presenter".
- **Max segment duration**: default 15 seconds. User can specify any value.
- **Narration audio duration**: if provided, used to calculate actual speech rate.
- **Speech rate**: if user directly specifies (e.g., "4 chars/second"), skip estimation.
- **Style direction**: user expresses freely (e.g., mature, casual, professional, playful, sensual-but-professional). Affects acting only, not appearance.
- **Aspect ratio**: user sets in video generation software. Skill does not hardcode.
- **Narration language**: default Chinese.

## Reference Files

This Skill includes three self-contained reference files:

| File | Purpose |
|------|---------|
| `references/ref2va-format-guide.md` | Complete Ref2VA six-section format specification with shot rules, camera vocabulary, dialogue tags, and continuity checklist |
| `references/narration-budget-guide.md` | Speech rate estimation, word counting rules, segment splitting algorithm, and validation criteria |
| `references/hologram-library.md` | 12 hologram types, content matching rules, interaction vocabulary, and color derivation rules |

Read the relevant reference file at each step as instructed below.

---

## STEP 1: Parse Input

### 1.1 Extract Character Identity (from Picture 1)

Describe the character with these mandatory attributes:
- Ethnicity and age category
- Hair: style, length, color, distinctive features (ribbon, bow, etc.)
- Eyewear: frame type, color, chain or strap
- Top clothing: color, style, neckline
- Accessories: jewelry, necklace, ribbon
- Makeup style: natural, bold, luxury

### 1.2 Extract Environment (from Picture 2)

Describe the environment with these mandatory attributes:
- Desk: type, color, shape
- Computer/device: type, color
- Lighting: lamp type, shade color, LED strips, ambient mood
- Chair: material, color, base
- Shelves: type, contents
- Flowers/plants: type, color
- Wall decor: framed prints, signs
- Overall color palette and ambiance

### 1.3 Parse Script

- Preserve the original script verbatim
- Count total characters using the rules in `references/narration-budget-guide.md`
- Identify content features: technical density, conversational markers, emotional markers

### 1.4 Determine Style

- Read the user's style description if provided
- Do NOT impose a fixed style. The user expresses freely.
- Style affects acting and narration delivery only, NOT character appearance.

---

## STEP 2: Calculate Narration Budget

Read `references/narration-budget-guide.md` for the complete guide.

### 2.1 Determine Speech Rate

Priority order:
1. If user provides audio duration: `rate = total_chars / audio_duration`
2. If user directly specifies rate: use that value
3. Otherwise: estimate from content density (see narration-budget-guide.md)

### 2.2 Calculate Per-Segment Budget

```
hologram_pause_total = sum of all hologram interaction pauses in the segment
                   (minimum 2 interactions, each 0.5-2.0 seconds based on complexity)
available_chars = speech_rate * (max_segment_duration - hologram_pause_total)
```

### 2.3 Output Budget Table

| Parameter | Value |
|-----------|-------|
| Total script chars | XXX |
| Speech rate | X.X chars/sec |
| Max segment duration | XX sec |
| Hologram pauses per segment | X.X sec |
| Available chars per segment | XX |

---

## STEP 3: Split Script into Segments

Read `references/narration-budget-guide.md` for the complete splitting algorithm.

From StarAI-2026/ComfyUI-MiniMax-H3-SkillBridge, MIT.

More from StarAI-2026/ComfyUI-MiniMax-H3-SkillBridge