Video Script

Extract from input:

• Core topic and key message
• Target audience and expertise level
• Desired video length (15s, 30s, 60s)
• Tone (educational, promotional, conversational, urgent)
• Key points that must be included
• Specific CTAs or desired outcomes
• Target platform (Sora 2, Veo 3.1, or both)

Step 2: Select Script Structure

Choose structure based on video length and AI platform constraints:

15-Second Script (Single Segment - Sora 2/Veo 3.1 Optimized)

Structure: Hook → Context → Value → CTA Target Word Count (English): 35-41 words (promotional pace, 160 WPM) Target Word Count (Spanish): 32-38 words (promotional pace, 145 WPM) Best for: Platform-native content, single-idea videos, maximum AI generation reliability

Critical: Both Sora 2 (15s max) and Veo 3.1 (max 8-second clips, but can batch 1-4 clips) benefit from 15-second segment planning.

30-Second Script (2 × 15s Segments)

Structure:

• Segment 1 (0-15s): Hook → Context
• Segment 2 (15-30s): Value → CTA

Word count: 70-82 words total (English promotional) Best for: Reels, TikTok, educational content, testimonials

60-Second Script (4 × 15s Segments)

Structure:

• Segment 1 (0-15s): Hook
• Segment 2 (15-30s): Problem + Context
• Segment 3 (30-45s): Solution Part 1
• Segment 4 (45-60s): Solution Part 2 + CTA

Word count: 140-164 words total (English promotional) Best for: Deep dives, case studies, detailed tutorials

Production Note: For Sora 2, generate each segment separately and stitch in post. For Veo 3.1, batch generate using Ingredients-to-Video for character continuity.

Step 3: Generate Script Components

Create each narrative component using proven patterns:

Hook (0-3 seconds)

Purpose: Stop the scroll, spark immediate curiosity

Pattern Library:

• Question pattern: "What if you could [desirable outcome]?"
• Mistake pattern: "Here's what you're doing wrong with [topic]..."
• Promise pattern: "3 steps to [specific result]"
• Curiosity pattern: "I bet you never knew this about [topic]..."
• Shock pattern: "Don't make this mistake when [action]..."
• Direct pattern: "Stop [bad behavior]. Start [good behavior]."

Deliverable: Generate 2-3 hook alternatives for A/B testing

See references/hook-library.md for 50+ proven hook templates.

Context/Problem (3-5 seconds)

Purpose: Establish relevance and show understanding

Pattern:

• State relatable situation: "If your [thing] isn't [desired state]..."
• Acknowledge struggle: "When you try to [goal] but [obstacle]..."
• Identify frustration: "Many [audience] hit a wall because [reason]..."

Key: Be specific and relatable, not generic.

Value/Solution (5-10 seconds)

Purpose: Deliver core insight or teaching

Pattern:

• Tease first: "Here's a [method] that changes everything."
• Deliver insight: "The key is [specific concept/formula]."
• Explain briefly: "This works because [reason]."

Constraints by Length:

• 15s videos: 1 main point only
• 30s videos: 1-2 main points maximum
• 60s videos: 2-3 points with brief explanations

CTA (2-4 seconds)

Purpose: Direct viewer to next action

Pattern Options (lowest to highest commitment):

• Engagement: "Save this for later"
• Comment: "Comment [word] below and I'll send you [thing]"
• DM: "DM me '[keyword]' for the [resource]"
• Link: "Click link in bio to get [resource]"
• Multi-step: "Like this, follow for more, and grab the free guide in bio"

Deliverable: Generate 2 CTA variations for testing

Step 4: Timing Validation

CRITICAL: Scripts must fit within target duration with buffer time.

Language-Specific Timing Guidelines

English:

• Conversational: 140-150 WPM = 35-38 words per 15s
• Professional: 150-160 WPM = 38-40 words per 15s
• Promotional: 160-170 WPM = 40-42 words per 15s

Spanish (Latin American):

• Conversational: 130-140 WPM = 32-35 words per 15s
• Professional: 140-150 WPM = 35-38 words per 15s
• Promotional: 145-155 WPM = 36-39 words per 15s

See references/timing-guidelines.md for 10+ languages with complete WPM tables.

Validation Process

1. Count words in final script
2. Calculate duration: (word count ÷ WPM) × 60 = seconds
3. Verify: Duration ≤ target segment duration - 0.75s buffer
4. Read aloud at intended energy level to confirm natural pacing
5. If over time: Reduce by 10-15% and retest

Buffer Requirements:

• 15s segment: 0.75s buffer (target 14.0-14.25s of speech)
• 30s video: 1.5s total buffer
• 60s video: 3.0s total buffer

Step 5: Add Production Elements

Gesture Cues

Map physical gestures to script moments for AI video generation:

Gesture Patterns by Moment:

• Opening (0-3s): Open palms, welcoming gesture
• Questions (any): Palms up, inviting response
• Key points: Counting fingers, pointing emphasis
• Emphasis: Two-handed gestures
• Closing: Forward lean, inviting gesture toward camera

Format:

0-3s → Open hand gesture (welcoming) — "[script phrase]"
3-7s → Pointing emphasis — "[script phrase]"
7-10s → Two-handed gesture — "[script phrase]"

Visual Overlay Suggestions

Provide B-roll, text overlay, and emphasis recommendations:

Overlay: 'Key Formula = Value' at 5s mark
B-roll: Show [specific visual] during "[key phrase]"
Pause: 0.3s before key number
Text emphasis: 'BENEFIT' at 8s

Pacing Markers

Indicate delivery adjustments:

[PAUSE 0.3s]
[EMPHASIS on "critical statistic"]
[SLOWER pace: "Here's what you do"]
[ENERGY UP: "Act now"]

Structured Output Format

CRITICAL: Output must follow this exact section structure for post-processing scripts and integration with video-prompt skill.

## SCRIPT METADATA

**Title:** [Descriptive video title]
**Total Duration:** [X] seconds
**Segments:** [N] × 15s clips
**Language:** [Language]
**Tone:** [Conversational/Professional/Promotional/Urgent]
**Platform Optimization:** [Sora 2 | Veo 3.1 | Both]
**Total Word Count:** [X] words

---

## HOOKS

### Hook Option A: [Hook Type Name]
**Pattern:** [Question/Mistake/Promise/etc.]
**Text:** "[Complete hook text]"
**Psychology:** [Why this hook works]

### Hook Option B: [Hook Type Name]
**Pattern:** [Hook type]
**Text:** "[Complete hook text]"
**Psychology:** [Why this hook works]

### Hook Option C: [Hook Type Name]
**Pattern:** [Hook type]
**Text:** "[Complete hook text]"
**Psychology:** [Why this hook works]

---

## SEGMENTS

### Segment 1 (0-15s) — [Narrative Component Names]

**Full Script:**
[Complete script text for this segment, optimized for timing]

**Word Count:** [X] words
**Calculated Duration:** [Y] seconds at [Z] WPM
**Validation:** ✓ Within 14.0-14.5s target

**Gesture Cues:**
* 0-3s → [Gesture description] — "[key phrase]"
* 3-7s → [Gesture description] — "[key phrase]"
* 7-10s → [Gesture description] — "[key phrase]"
* 10-15s → [Gesture description] — "[key phrase]"

**Visual Suggestions:**
* [Overlay/B-roll/emphasis recommendation]
* [Text overlay timing and content]
* [Pacing notes]

---

### Segment 2 (15-30s) — [Narrative Component Names]

[Same format as Segment 1]

---

[Additional segments for 30s/60s videos]

---

## CTA VARIATIONS

### CTA Option 1: [CTA Type — Engagement/Comment/DM/Link]
**Text:** "[Complete CTA text]"
**Commitment Level:** [Low/Medium/High]
**Best For:** [Content type/audience context]

### CTA Option 2: [CTA Type]
**Text:** "[Complete CTA text]"
**Commitment Level:** [Level]
**Best For:** [Context]

---

## PRODUCTION NOTES

### Narrative Structure
[Overall arc description: Hook → Context → Value → CTA]

### Energy & Pacing
[Delivery style notes, energy modulation across segments]

### Platform-Specific Considerations
[Sora 2: Generate segments separately, stitch in post]
[Veo 3.1: Use Ingredients-to-Video for character continuity]

### Post-Production Requirements
[Any compositing, editing, or stitching requirements]

### Alternative Structures
[If applicable, show alternative ways to structure the same content]

Post-processing scripts parse these exact section headers (## SCRIPT METADATA, ## HOOKS, ## SEGMENTS, ## CTA VARIATIONS, ## PRODUCTION NOTES) to extract structured data for JSON export or direct integration with video-prompt generation.

Multi-Language Support

Timing Tables by Language

See references/timing-guidelines.md for complete multi-language timing specifications.

Integration with video-prompt Skill

This skill outputs timed, structured scripts. The video-prompt skill adds production specifications.

This skill provides:

• Timed script segments (validated to 15s constraints)
• Hook alternatives
• CTA variations
• Gesture cues
• Visual overlay suggestions
• Narrative structure

Next skill (video-prompt) adds:

• Character specifications (@username, wardrobe, appearance)
• Camera setup (lens, angle, framing, movement)
• Lighting details (3-point lighting, color temp, modifiers)
• Background specifications (chroma key, location, solid color)
• Complete platform prompts (Sora 2 or Veo 3.1 format)

Combined Workflow:

Raw idea/concept
  ↓
video-script (this skill)
  ↓
Timed, segmented script with narrative structure
  ↓
video-prompt
  ↓
Production-ready AI video prompts (platform-specific)
  ↓
Generate in Sora 2 or Veo 3.1 → Stitch if needed

Quality Checklist

Before finalizing script, verify:

Content:

• Clear hook that stops the scroll (2-3 alternatives provided)
• Relatable problem/context established
• Core value/solution communicated clearly
• Strong CTA with clear action (2 variations provided)
• Narrative arc appropriate for video length

Timing:

• Word count within target range for language
• Read-aloud test confirms duration fits segment
• Buffer time included (0.75s per 15s segment)
• Natural pacing (not rushed or overly slow)
• All segments independently validated

Production Readiness:

• Gesture cues mapped to specific moments
• Visual suggestions included per segment
• Tone clearly specified
• Energy modulation noted where needed
• Platform optimization specified (Sora 2/Veo 3.1/Both)

Structure:

• All required section headers present
• Segment breakdown clear and labeled
• Metadata complete
• Alternative hooks and CTAs provided

Language:

• Conversational and natural
• No jargon (unless audience-appropriate)
• Active voice used
• Specific, not generic

Common Patterns

Pattern: Educational Content (15s, Single Segment)

Structure: Hook → Context → Value → CTA

Example (English, 160 WPM):

Hook: "Here's what you need to know about DSCR loans."
Context: "Most people miss this key detail."
Value: "The secret is rental income, not W-2s. If your property brings in 1.2 times the payment, you qualify."
CTA: "Save this for later."

Total: 38 words = 14.25s at 160 WPM ✓

Pattern: Promotional Campaign (30s, 2 Segments)

Segment 1 (Hook + Context): Grab attention, establish offer or problem

Segment 2 (Value + CTA): Deliver benefit, urgency, and action

Pattern: Tutorial/Deep Dive (60s, 4 Segments)

Segment 1: Hook Segment 2: Problem + Context in detail Segment 3: Solution Part 1 (core concept) Segment 4: Solution Part 2 + Implementation + CTA

Troubleshooting

Issue: Script feels rushed when read aloud

Solution:

• Reduce word count by 15-20%
• Remove filler words ("actually," "basically," "just")
• Simplify complex phrases
• Add [PAUSE] markers where needed
• Adjust WPM assumption (use conversational instead of promotional)

Issue: Hook not compelling enough

Solution:

• Check references/hook-library.md for proven patterns
• Try different psychological trigger (question → shock → promise)
• Make it more specific to target audience
• Lead with surprising insight or statistic
• Test with curiosity gap pattern

Issue: Script too generic or boring

Solution:

• Add specific numbers, examples, or names
• Use "you" language (second person)
• Include concrete outcomes, not abstractions
• Add personality or unique perspective
• Replace general statements with specific benefits

Issue: CTA feels weak or unclear

Solution:

• Be more specific about the action
• Add value proposition ("get the [specific resource]")
• Create urgency if appropriate ("first 10 people...")
• Make it one clear action (not multiple competing CTAs)
• Match CTA commitment level to content value

Issue: Doesn't fit target segment duration

Solution:

• Count words vs. target timing table
• If over: Cut least essential element, remove redundancy
• If under: Expand problem or solution with specific details
• Test by reading aloud at intended pace and energy
• Adjust WPM category if delivery style changes

Example Output

30-Second Real Estate Script (English, 2 × 15s Segments)

## SCRIPT METADATA

**Title:** DSCR Loans Explained for Real Estate Investors
**Total Duration:** 30 seconds
**Segments:** 2 × 15s clips
**Language:** English
**Tone:** Educational/Professional
**Platform Optimization:** Both (Sora 2 + Veo 3.1)
**Total Word Count:** 78 words

---

## HOOKS

### Hook Option A: Question Pattern
**Pattern:** Question
**Text:** "What if you could get a loan based on your property's cash flow—not your paycheck?"
**Psychology:** Engages curiosity, promises solution to common pain point

### Hook Option B: Mistake Pattern
**Pattern:** Mistake/Problem
**Text:** "Most investors hit a wall because they can't qualify using W-2s."
**Psychology:** Identifies relatable frustration, positions solution

### Hook Option C: Promise Pattern
**Pattern:** Promise/Benefit
**Text:** "Property loans without W-2s? Here's exactly how."
**Psychology:** Direct benefit promise, creates knowledge gap

---

## SEGMENTS

### Segment 1 (0-15s) — Hook + Context

**Full Script:**
What if you could get a loan based on your property's cash flow—not your paycheck? Many investors hit a wall because they don't qualify using W-2s. Here's a DSCR method that changes that.

**Word Count:** 38 words
**Calculated Duration:** 14.25s at 160 WPM
**Validation:** ✓ Within 14.0-14.5s target

**Gesture Cues:**
* 0-3s → Open palms (welcoming) — "What if you could..."
* 3-8s → Pointing gesture (emphasis) — "Many investors hit a wall"
* 8-15s → Forward lean (engaging) — "Here's a DSCR method"

**Visual Suggestions:**
* B-roll: Property exterior or cash flow diagram
* Text overlay: "Cash Flow > Paycheck" at 7s
* Pause 0.3s after "W-2s" for emphasis

---

### Segment 2 (15-30s) — Value + CTA

**Full Script:**
DSCR means your rental income covers 1.2 times your payment. Lenders see you as low risk. I've done 100+ of these. If the numbers work, you qualify. DM me 'DSCR' for my free tool.

**Word Count:** 40 words
**Calculated Duration:** 15.0s at 160 WPM
**Validation:** ✓ Within 14.0-15.0s target (acceptable for final segment)

**Gesture Cues:**
* 15-18s → Counting gesture (one hand) — "1.2 times"
* 18-22s → Confident gesture (hands together) — "low risk"
* 22-26s → Pointing to camera — "I've done 100+"
* 26-30s → Inviting gesture toward camera — "DM me 'DSCR'"

**Visual Suggestions:**
* Overlay: "DSCR = Rental Income ÷ Debt" at 16s
* B-roll: Calculator or spreadsheet visual
* Text emphasis: "100+ Deals" at 24s
* Stronger energy for CTA

---

## CTA VARIATIONS

### CTA Option 1: DM for Resource
**Text:** "DM me 'DSCR' for my free tool."
**Commitment Level:** Medium
**Best For:** Educational content with lead magnet offer

### CTA Option 2: Save for Later
**Text:** "Save this and follow for more investor tips."
**Commitment Level:** Low
**Best For:** Building audience, lower-commitment ask

---

## PRODUCTION NOTES

### Narrative Structure
Hook → Context → Problem → Solution → Credibility → CTA

### Energy & Pacing
- Segment 1: Moderate energy, build curiosity
- Segment 2: Confident energy, authoritative delivery on CTA

### Platform-Specific Considerations
**Sora 2:** Generate two separate 15s clips, stitch in post with cut or dissolve transition
**Veo 3.1:** Use Ingredients-to-Video mode with character ingredient for continuity across both clips

### Post-Production Requirements
- Stitch segments with 0.5s crossfade or hard cut
- Add overlays at specified timestamps
- Ensure audio levels consistent across segments

Reference Files

This skill includes detailed reference materials:

• timing-guidelines.md - Multi-language WPM tables, buffer calculations, timing formulas for 10+ languages
• hook-library.md - 50+ proven hook templates across psychological trigger types (curiosity, fear, desire, urgency, etc.)

Refer to these files for deeper guidance on specific components.