Scenario Prompt 
For Midjourney and Google Veo 3 Video 

**TASK:**  

Create a full scenario-based training module focused on customer support for a new software upgrade.  

The deliverables must include:  

1. One realistic training scenario between a support agent and a customer  

2. Five MidJourney image prompts with visual consistency (same characters and environment)  

3. A short action prompt and spoken script for each image that will be used in VEO3 to create 8-second video clips  

4. Each spoken script must sound natural, fit within 8 seconds, and include realistic tone and pacing

**ROLE:**  

Act as a training developer, scenario writer, and MidJourney prompt engineer. Ensure instructional quality, cinematic visual design, and realistic dialogue.

**AUDIENCE:**  

This training is for customer support professionals learning how to handle calls about a recent software update. Learners need to see good practices modeled.

**CREATE:**  

The customer is frustrated that a feature they use daily has been removed in a recent software upgrade. The support agent must calm the customer, validate their frustration, and walk them through the new way to perform the same task.

Provide the output in the following structure:

### 🟦 Scenario Title:  

[Title of the scenario]

### 🟪 Background:  

[Brief situation summary]

### 🟩 Characters:  

- **Support Agent:** [description]  

- **Customer:** [description]

### 🟨 Scenario Objective:  

[Clear skill the learner should take away]

## ⬛ Scene Breakdown

For each of the following scenes, provide:

- **Scene Name:**  

- **MidJourney Image Prompt:**  

- **VO3 Action Prompt:**  

- **VO3 Spoken Script:**  

Create 5 total scenes:  

1. Frustrated Customer (opening)  

2. Empathetic Agent Response  

3. Customer Objection / Doubt  

4. Agent Showing the Fix  

5. Customer Accepting the Solution

Style: Photorealistic, cinematic, 16:9 aspect, consistent lighting, expressions, and clothing. Use `--style raw` and `--v 7` in MidJourney and keep characters visually consistent using a seed value.

**INTENT:**  

Model effective emotional intelligence, product knowledge, and user guidance. Create both the visual and verbal assets for an interactive training scenario.

Interactive Video Prompt 
For Google Gemini 2.5 Pro

**TASK:**  
Create a responsive HTML/JavaScript/CSS interface that overlays interactive questions on top of a single .mp4 video containing 5 scenario-based scenes. The questions appear at defined timestamps (default break at 8 seconds, but adjustable) and pause the video for user interaction.

**ROLE:**  
You are a senior front-end developer experienced in multimedia interactivity, JavaScript video APIs, and responsive design for educational content delivery.

**AUDIENCE:**  
Corporate learners consuming scenario-based video training content. They interact with embedded questions between scene transitions to reinforce decision-making and assess understanding.

**CREATE:**  
Generate clean, modular HTML/JavaScript/CSS code that:

• Loads and plays a single MP4 file with a visible video player
• Pauses at specified timestamps (e.g., every 8s or custom)
• Overlays a full-screen translucent modal with a question and multiple-choice answers
• Resumes the video when the correct choice is made (or on button click if there's no validation logic)
• Works smoothly on desktop and mobile
• Allows timecode breakpoints to be easily updated in an array or JSON structure
• Ensures questions are specific to the scene (e.g., 5 scenes → 5 questions)
• Is fully self-contained (no external frameworks like React or jQuery unless explicitly allowed)

Example Timecode Input Format:

const breakpoints = [

  { time: 8, question: "What should the manager do next?", options: ["Option A", "Option B"], correct: 1 },

  { time: 16, question: "How should the team respond?", options: ["Option A", "Option B", "Option C"], correct: 2 },

  { time: 24, question: "What is the root cause?", options: ["Option A", "Option B"], correct: 0 },

  { time: 32, question: "Which policy applies?", options: ["Policy A", "Policy B", "Policy C"], correct: 1 },

  { time: 40, question: "How should feedback be delivered?", options: ["Option A", "Option B"], correct: 0 }

];

Output Requirements:

1. HTML:

• A container with a <video> element
• A hidden modal/question overlay with dynamic content placeholders

2. CSS:

• Responsive layout
• Fullscreen translucent overlay
• Styled modal box and buttons

3. JavaScript:

• timeupdate event listener to track video progress
• Comparison logic to match current time with breakpoints
• Pause logic and overlay trigger
• Dynamic rendering of questions/options
• Input handler for answer selection
• Resume logic to continue playback
• Easy-to-edit breakpoints array for customization

**INTENT:**  
To create reusable training experiences that mix passive watching with active participation, using video-based scenarios. This supports better learner engagement, knowledge checks, and adaptive feedback inside one seamless video interface.