This skill should be used when the user asks to "generate video prompts", "create Sora prompts", "create Veo prompts", "write AI video prompts", "generate text-to-video prompts", "create character sheets for video", "optimize prompts for Sora 2", "optimize prompts for Veo 3.1", "create avatar video backgrounds", or needs platform-specific AI video generation prompts with detailed production specifications.
From video-production-suitenpx claudepluginhub nbkm8y5/claude-plugins --plugin video-production-suiteThis skill uses the workspace's default tool permissions.
references/avatar-backgrounds.mdreferences/real-estate-content.mdreferences/sora2-syntax.mdreferences/south-florida-locations.mdreferences/veo31-syntax.mdExecutes pre-written implementation plans: critically reviews, follows bite-sized steps exactly, runs verifications, tracks progress with checkpoints, uses git worktrees, stops on blockers.
Guides idea refinement into designs: explores context, asks questions one-by-one, proposes approaches, presents sections for approval, writes/review specs before coding.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
Generate platform-specific AI video prompts from scripts or concepts for Sora 2 and Veo 3.1. This skill transforms timed video scripts into production-ready text-to-video prompts with complete character specifications, lighting setups, camera controls, and background details optimized for each platform's capabilities.
This skill creates comprehensive video generation prompts for two leading AI video platforms:
The skill handles character continuity across multi-segment videos, platform-specific formatting, cost optimization, and integration with avatar/B-roll background libraries.
Platform Selection (--platform flag):
sora — Sora 2 onlyveo — Veo 3.1 onlyboth (default) — Generate prompts for both platformsSora 2 Generation Modes:
Veo 3.1 Generation Modes:
Collect these essential details before generating prompts:
Platform Differences:
For multi-segment videos requiring character continuity, create a reusable CHARACTER SHEET:
## CHARACTER SHEET
**Identity:** @username (Sora Cameo) | [Full physical description for Veo]
**Physical Appearance:**
- Age: [Specific age or range]
- Ethnicity: [Specific description]
- Gender: [Male/Female/Non-binary]
- Build: [Athletic, slim, average, stocky, etc.]
- Height reference: [Tall, average, short — for framing context]
**Wardrobe:**
- Top: [Exact description with color — e.g., "navy blue collared polo shirt"]
- Bottom: [If visible — e.g., "dark grey trousers"]
- Accessories: [Watch, jewelry, glasses — be specific]
- Style notes: [Professional, casual, business casual, etc.]
**Hair & Styling:**
- Length: [Short, medium, long, shoulder-length, etc.]
- Color: [Specific color]
- Style: [Straight, wavy, curly, slicked back, etc.]
- Facial hair: [Clean-shaven, stubble, beard, mustache — describe]
**Default Expression & Demeanor:**
- Facial expression: [Friendly smile, serious, confident, approachable]
- Eye contact: [Direct to camera, natural, engaging]
- Posture: [Upright, relaxed, confident]
- Energy level: [Energetic, calm, professional, warm]
**Gesture Style:**
- Hand gestures: [Natural, expressive, minimal, emphatic]
- Movement quality: [Organic, professional, animated, composed]
- Specific preferences: [Avoid/include specific gestures]
**Platform-Specific:**
- Sora 2 Cameo: @[username]
- Veo 3.1 Ingredient: [Save character from first generation as ingredient]
This character sheet is copied verbatim into each segment prompt to ensure continuity.
Create complete prompts for each 15-second segment (or shorter for Veo batching).
[Style/Era/Aesthetic] — [Subject with complete character description] [Action with timing beats] in [Setting with lighting details]. [Camera specifications]. [Background description].
[Subject pronoun] speaks in [language/dialect], saying: "[exact script text]". [Gesture cues with timing]. [Energy/delivery notes]. [Final action/expression].
Sora 2 Prompt Template:
A [age range] [ethnicity] [gender], @[username], [energy descriptor] and [demeanor descriptor], stands facing the camera at [camera angle]. The camera uses a [focal length] lens showing [framing description]. [Pronoun] wears [complete wardrobe description], conveying [intended impression]. [Pronoun] has [hair/styling details] and a [expression description].
The scene takes place in a [location type] with [lighting quality] at [color temperature]:
– A [key light type and modifier] positioned [position description], casting [light quality].
– A [fill light type] on the [position] to [purpose].
– A [back light description] to [purpose].
The background is a [background description with specific technical requirements].
@[username] speaks naturally in [language/dialect], maintaining [energy descriptor]. [Pronoun] emphasizes [key content points] — "[direct script quote]" — while using [gesture description with timing]. [Pronoun] delivery is [delivery style descriptors].
The camera remains [movement description] at [position] with [movement notes]. The scene ends with [pronoun] [final action/expression], [final gesture description] — [intended final impression].
[Subject with complete physical description] [Action with specific verbs] [Environment with lighting]. [Camera framing and angle]. [Background description].
Veo 3.1 Prompt Template:
[Age range] [ethnicity] [gender] with [physical details], wearing [complete wardrobe], [action verb] while speaking to camera. [Pronoun] has [hair/styling] and [expression]. [Pronoun] uses [gesture description] as [pronoun] delivers: "[script text]".
Setting: [Location type]. Lighting: [key light position and quality], [fill light], [back light]. [Color temperature]. Background: [specific background description].
Camera: [Framing] at [angle]. [Camera preset: Static/Pan/Dolly/etc.]
Energy: [Energy descriptor]. Delivery: [Pacing/tone notes].
Veo Camera Presets (use instead of describing movement):
For videos longer than 15 seconds, create independent prompts for each segment that can be generated separately and stitched.
Critical Continuity Elements (must be identical across all segments):
Never reference previous segments — each prompt must be fully self-contained.
CRITICAL: Output must follow this exact section structure for post-processing scripts and integration with video generation workflows.
## PROMPT METADATA
**Project Title:** [Descriptive title]
**Total Duration:** [X] seconds
**Segments:** [N] × 15s (Sora 2) | [M] × 8s batches (Veo 3.1)
**Platform:** [Sora 2 | Veo 3.1 | Both]
**Generation Mode:** [Text-to-Video | Image-to-Video | Storyboard | etc.]
**Language:** [Script language]
---
## CHARACTER SHEET
[Complete character specifications as described above]
---
## SEGMENTS
### Segment 1 (0-15s) — [Narrative Component]
#### Sora 2 Prompt
**Mode:** Text to Video
**Duration:** 15s
**Model:** sora-2-pro
**Orientation:** Portrait (9:16) | Landscape (16:9)
**Prompt:**
[Complete Sora 2 prompt following template above]
**Script Text:**
"[Exact script for this segment]"
**Gesture Cues:**
* 0-3s → [Gesture] — "[key phrase]"
* 3-7s → [Gesture] — "[key phrase]"
* 7-10s → [Gesture] — "[key phrase]"
* 10-15s → [Gesture] — "[key phrase]"
---
#### Veo 3.1 Prompt
**Mode:** Text to Video (or Ingredients if character exists)
**Camera Control:** [Preset name]
**Aspect Ratio:** [16:9 | 9:16 | 1:1]
**Prompt:**
[Complete Veo 3.1 prompt following template above]
**Script Text:**
"[Exact script for this segment]"
**Gesture Cues:**
* 0-3s → [Gesture] — "[key phrase]"
* 3-7s → [Gesture] — "[key phrase]"
* 7-10s → [Gesture] — "[key phrase]"
* 10-15s → [Gesture] — "[key phrase]"
---
### Segment 2 (15-30s) — [Narrative Component]
[Same format as Segment 1, with identical character/lighting/background specs]
---
[Additional segments as needed]
---
## GENERATION SETTINGS
### Sora 2 Configuration
- **Model:** sora-2 (fast iteration) | sora-2-pro (production quality)
- **Orientation:** Portrait 9:16 | Landscape 16:9
- **Duration per clip:** 10s (1 credit) | 15s (2 credits) | 25s storyboard (4 credits, Pro only)
- **Cameo character:** @[username] (if applicable)
- **Audio:** Auto-generated by Sora
### Veo 3.1 Configuration
- **Model:** Veo 3.1 Fast
- **Aspect Ratio:** 16:9 | 9:16 | 1:1
- **Generation strategy:** Single clip | Batch 2-4 clips
- **Character ingredient:** [Save from first generation if multi-segment]
- **Camera presets per segment:** [List preset for each]
### Post-Production Requirements
- **Stitching:** [Hard cut | Crossfade | Transition type] between segments
- **Audio:** [Voiceover sync | Auto-generated | Music overlay]
- **Compositing:** [Chroma key replacement | Background swap | None]
- **Color grading:** [Consistency notes across segments]
---
## COST ESTIMATE
### Sora 2 (Pro Subscription: $200/month)
- Segment 1: [10s = 1 credit | 15s = 2 credits | 25s = 4 credits]
- Segment 2: [Credits]
- [Additional segments]
- **Total credits:** [X]
- **Estimated generations:** [Assuming 2-3 attempts per segment]
### Veo 3.1 (Labs Access: Free/Paid tiers vary)
- Segment 1: [1 clip | Batch of 4]
- Segment 2: [Configuration]
- [Additional segments]
- **Total clips:** [X]
- **Batch strategy:** [Single generation with Ingredients | Multiple text-to-video]
### Cost Optimization Tips
- Use sora-2 (not Pro) for iterative testing
- Generate Sora 10s clips instead of 15s where possible (50% fewer credits)
- For Veo, batch segments using Ingredients-to-Video for character continuity
- Test prompts with shorter durations before committing to full segments
For talking-head videos requiring clean, motion-friendly backgrounds:
Use case: Maximum post-production flexibility
Sora 2:
The background is a perfectly uniform solid green chroma screen (#00FF00) with no shadows, gradients, or texture, ideal for post-production compositing. The lighting is even across the entire background with no hotspots or variations.
Veo 3.1:
Background: Uniform solid green (#00FF00) chroma key screen, perfectly even lighting, no texture or shadows.
Use case: Corporate, educational, authoritative content
Sora 2:
The background is a softly blurred modern office interior with warm ambient lighting. Subtle bookshelves and neutral decor are visible but out of focus, creating depth without distraction. Soft natural light from an off-camera window adds warmth.
Veo 3.1:
Background: Soft-focus modern office with bookshelves, neutral decor, warm ambient lighting, shallow depth of field.
Use case: Contemporary, design-focused, lifestyle content
Sora 2:
The background is a clean, minimalist interior with soft white walls and subtle texture. Gentle indirect lighting creates a fresh, airy feel. No distinct objects or text visible, maintaining focus on the subject.
Veo 3.1:
Background: Minimal white textured wall, soft indirect lighting, clean aesthetic, no distractions.
Use case: Personal brand, approachable, conversational content
Sora 2:
The background features a softly blurred warm interior with plants and natural elements visible but unfocused. Golden hour lighting creates a welcoming, authentic atmosphere. Depth without visual clutter.
Veo 3.1:
Background: Soft-focus warm interior, plants, natural elements, golden hour lighting, shallow DOF.
Background Best Practices:
| Feature | Sora 2 | Veo 3.1 |
|---|---|---|
| Max clip duration | 15s standard, 25s Storyboard (Pro) | 8s base, batch up to 4 clips |
| Character continuity | Cameo (@username) | Ingredients-to-Video |
| Camera control | Prose description | Preset controls |
| Prompt style | Detailed prose, narrative | Structured, component-based |
| Multi-scene | Storyboard mode (Pro) | Scene Builder timeline |
| Image input | Image-to-Video mode | Frames-to-Video mode |
| Audio | Auto-generated with prompt cues | Silent (add in post) |
| Cost model | $200/month Pro subscription | Labs access (varies) |
| Best for | Narrative content, character-driven | Clean execution, batch generation |
For 30-60s videos, use Storyboard mode to generate connected segments:
Limitations: 25s total max, Pro subscription required
Alternative: Generate segments individually and stitch in post-production
For multi-segment videos requiring character continuity:
Benefits: Reliable character continuity, batch 1-4 clips per generation
Before finalizing prompts, verify:
Prompt Completeness:
Character Continuity (Multi-Segment):
Platform Optimization:
Production Readiness:
Script Integration:
This skill receives timed scripts from video-script and adds production specifications.
Input from video-script:
This skill adds:
Combined Workflow:
Raw idea/concept
↓
video-script skill
↓
Timed, segmented script with narrative structure
↓
video-prompt skill (this skill)
↓
Platform-specific AI video prompts (Sora 2 and/or Veo 3.1)
↓
Generate in platform → Stitch segments → Final video
Solution:
Solution:
Solution:
Solution:
Solution:
Solution:
## PROMPT METADATA
**Project Title:** October DSCR Loan Campaign - UBI Equity Group
**Total Duration:** 30 seconds
**Segments:** 2 × 15s (Sora 2) | 4 × 8s batches (Veo 3.1)
**Platform:** Both (Sora 2 + Veo 3.1)
**Generation Mode:** Text-to-Video (Sora) | Ingredients-to-Video (Veo, Segment 2+)
**Language:** English
---
## CHARACTER SHEET
**Identity:** @carlosmoreno (Sora Cameo) | Carlos Moreno, Latino real estate professional
**Physical Appearance:**
- Age: Early 40s
- Ethnicity: Latino (Colombian descent)
- Gender: Male
- Build: Athletic, fit
- Height reference: Average height, confident posture
**Wardrobe:**
- Top: Navy blue collared polo shirt with subtle UBI Equity Group logo on left chest
- Style: Business casual, professional, approachable
**Hair & Styling:**
- Length: Short, neatly trimmed
- Color: Dark brown with slight grey at temples
- Style: Clean, professional cut
- Facial hair: Light stubble, well-groomed
**Default Expression & Demeanor:**
- Facial expression: Confident, friendly smile
- Eye contact: Direct to camera, engaging and warm
- Posture: Upright, relaxed confidence
- Energy level: Professional yet approachable, energetic
**Gesture Style:**
- Hand gestures: Natural, professional, emphatic at key points
- Movement quality: Organic, confident, not over-rehearsed
- Specific preferences: Open palm gestures for welcoming, pointing for emphasis
**Platform-Specific:**
- Sora 2 Cameo: @carlosmoreno
- Veo 3.1 Ingredient: Save from Segment 1 generation as "Carlos - UBI Professional"
---
## SEGMENTS
### Segment 1 (0-15s) — Hook + Context
#### Sora 2 Prompt
**Mode:** Text to Video
**Duration:** 15s
**Model:** sora-2-pro
**Orientation:** Portrait (9:16)
**Prompt:**
A Latino man in his early 40s, @carlosmoreno, confident and approachable, stands facing the camera at eye level. The camera uses a 35mm lens showing a waist-up frame. He wears a navy blue collared polo shirt with a subtle UBI Equity Group logo on the left chest, conveying professional credibility. He has short dark brown hair with slight grey at the temples, light well-groomed stubble, and a friendly, confident smile.
The scene takes place in a modern office setting with professional lighting at 4400K:
– A rectangular softbox positioned 45° overhead and to the left, casting soft, even light across the subject's face.
– A tube light on the right side to fill shadows and maintain natural skin tones.
– A back light to separate the subject from the background and add dimension.
The background is a softly blurred modern office interior with neutral decor and bookshelves visible but out of focus, creating depth without distraction.
@carlosmoreno speaks naturally in English, maintaining energetic enthusiasm. He emphasizes property investment opportunities — "What if you could get a loan based on your property's cash flow—not your paycheck?" — while using open palm gestures to invite engagement. His delivery is confident, warm, and educational.
The camera remains static at eye level. The scene ends with him leaning slightly forward with an inviting gesture toward the camera — creating connection and curiosity.
**Script Text:**
"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."
**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"
---
#### Veo 3.1 Prompt
**Mode:** Text to Video
**Camera Control:** Static
**Aspect Ratio:** 9:16
**Prompt:**
Latino man in early 40s with short dark hair, light stubble, athletic build, wearing navy blue polo with UBI Equity Group logo, speaks directly to camera with confident smile. He uses open palm gestures and pointing emphasis while delivering: "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."
Setting: Modern office interior. Lighting: Rectangular softbox 45° left, tube fill light right, back light for separation, 4400K color temperature. Background: Soft-focus office with bookshelves, neutral decor, shallow depth of field.
Camera: Waist-up framing at eye level. Static.
Energy: Professional, energetic, warm. Delivery: Confident and educational tone.
**Script Text:**
"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."
**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"
**Note:** Save this generation as Character Ingredient "Carlos - UBI Professional" for Segment 2 continuity
---
### Segment 2 (15-30s) — Value + CTA
#### Sora 2 Prompt
**Mode:** Text to Video
**Duration:** 15s
**Model:** sora-2-pro
**Orientation:** Portrait (9:16)
**Prompt:**
A Latino man in his early 40s, @carlosmoreno, confident and approachable, stands facing the camera at eye level. The camera uses a 35mm lens showing a waist-up frame. He wears a navy blue collared polo shirt with a subtle UBI Equity Group logo on the left chest, conveying professional credibility. He has short dark brown hair with slight grey at the temples, light well-groomed stubble, and a confident, authoritative expression.
The scene takes place in a modern office setting with professional lighting at 4400K:
– A rectangular softbox positioned 45° overhead and to the left, casting soft, even light across the subject's face.
– A tube light on the right side to fill shadows and maintain natural skin tones.
– A back light to separate the subject from the background and add dimension.
The background is a softly blurred modern office interior with neutral decor and bookshelves visible but out of focus, creating depth without distraction.
@carlosmoreno speaks naturally in English, maintaining confident authority. He emphasizes the DSCR solution and credibility — "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." — while using counting gestures and pointing toward the camera for the call to action. His delivery is assured, direct, and action-oriented.
The camera remains static at eye level. The scene ends with him using an inviting gesture toward the camera during the CTA — creating clear next step and urgency.
**Script Text:**
"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."
**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'"
---
#### Veo 3.1 Prompt
**Mode:** Ingredients to Video (using saved character from Segment 1)
**Camera Control:** Static
**Aspect Ratio:** 9:16
**Prompt:**
[Use saved character ingredient "Carlos - UBI Professional"]
Carlos delivers with confident authority: "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." He uses counting gestures at "1.2 times," confident hand position at "low risk," pointing at camera for "100+ of these," and inviting gesture toward camera for the DM call to action.
Setting: Modern office interior (same as Segment 1). Lighting: Rectangular softbox 45° left, tube fill light right, back light for separation, 4400K color temperature. Background: Soft-focus office with bookshelves, neutral decor, shallow depth of field.
Camera: Waist-up framing at eye level. Static.
Energy: Confident, authoritative, action-oriented. Delivery: Direct and assured.
**Script Text:**
"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."
**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'"
---
## GENERATION SETTINGS
### Sora 2 Configuration
- **Model:** sora-2-pro (production quality for campaign)
- **Orientation:** Portrait 9:16 (social media optimized)
- **Duration per clip:** 15s (2 credits each)
- **Cameo character:** @carlosmoreno (ensures character continuity)
- **Audio:** Auto-generated from script cues
### Veo 3.1 Configuration
- **Model:** Veo 3.1 Fast
- **Aspect Ratio:** 9:16 (vertical)
- **Generation strategy:** Segment 1 = Text-to-Video, save character as Ingredient; Segment 2 = Ingredients-to-Video with character
- **Character ingredient:** "Carlos - UBI Professional" (saved from Segment 1)
- **Camera preset:** Static (both segments)
### Post-Production Requirements
- **Stitching:** Hard cut or 0.5s crossfade between Segment 1 and Segment 2
- **Audio:** Sync professional voiceover or use AI-generated audio (Sora)
- **Overlays:** Add text overlays at specified timestamps:
- "Cash Flow > Paycheck" at 7s (Segment 1)
- "DSCR = Rental Income ÷ Debt" at 16s (Segment 2)
- "100+ Deals" at 24s (Segment 2)
- **Color grading:** Ensure consistency across segments, warm professional tone
---
## COST ESTIMATE
### Sora 2 (Pro Subscription: $200/month)
- Segment 1 (15s): 2 credits
- Segment 2 (15s): 2 credits
- **Total credits:** 4 credits per complete generation
- **Estimated generations:** 2-3 attempts per segment = 8-12 credits total
- **Monthly allowance:** Verify current Pro tier credit limit
### Veo 3.1 (Labs Access: Varies by tier)
- Segment 1 (Text-to-Video): 1 generation
- Segment 2 (Ingredients-to-Video): 1 generation
- **Total clips:** 2 clips
- **Batch strategy:** Generate individually, character continuity via Ingredients feature
- **Cost:** Depends on Labs tier (free tier: limited generations/day; paid: varies)
### Cost Optimization Tips
- Test with sora-2 (standard model) before committing to Pro for final generation
- For Veo: Ingredients-to-Video significantly improves character continuity, reducing re-generation attempts
- Generate 10s clips for testing if 15s not critical (saves 50% Sora credits)
- Batch multiple campaign variations using same CHARACTER SHEET to amortize setup work