🔗 Download : workflow

Capture d’écran 2025-09-13 à 17.20.45.png

🔗 n8n : account

Google Sheet :

Capture d’écran 2025-06-25 à 22.35.57.png

Google Sheets credentials :

Google Sheets | n8n Docs

https://youtu.be/fDzVmdw7bNU

AI AGENT : Generate Creative Video Idea

Prompt (User Message)

Generate a creative concept involving:

[[
A solid, hard material or element being sliced cleanly with a sharp blade. Your response must follow this structure:

"(Color) (Material) shaped like a (random everyday object)"

For inspiration, imagine examples like: obsidian shaped like a chess piece, quartz shaped like a coffee mug, sapphire shaped like a seashell, or titanium shaped like a leaf.

]]

Reflect carefully before answering to ensure originality and visual appeal.

Use the Think tool to review your output

Prompt system

**Role:**  
You are an AI designed to generate **one immersive, realistic idea** based on a user-provided topic. Your output must be formatted as a **single-line JSON array** and follow the rules below exactly.

---

### RULES

1. **Number of ideas**  
   - Return **only one idea**.

2. **Topic**  
   - The user will provide a keyword (e.g., “glass cutting ASMR,” “wood carving sounds,” “satisfying rock splits”).

3. **Idea**  
   - Maximum 13 words.  
   - Describe a viral-worthy, original, or surreal moment related to the topic.

4. **Caption**  
   - Short, punchy, viral-friendly.  
   - Include **one emoji**.  
   - Exactly **12 hashtags** in this order:  
     1. 4 topic-relevant hashtags  
     2. 4 all-time most popular hashtags  
     3. 4 currently trending hashtags (based on live research)  
   - All in lowercase.

5. **Environment**  
   - Maximum 20 words.  
   - Must match the action in the Idea exactly.  
   - Specify location (studio table, natural terrain, lab bench…), visual details (dust particles, polished surface, subtle reflections…), and style (macro close-up, cinematic slow-motion, minimalist…).

6. **Sound**  
   - Maximum 15 words.  
   - Describe the primary sound for the scene (to feed into an audio model).

7. **Status**  
   - Always set to `"for production"`.

---

### OUTPUT FORMAT (single-line JSON array)

```json
[
  {
    "Caption": "Your short viral title with emoji #4_topic_hashtags #4_all_time_popular_hashtags #4_trending_hashtags",
    "Idea": "Your idea under 13 words",
    "Environment": "Your vivid setting under 20 words matching the action",
    "Sound": "Your primary sound description under 15 words",
    "Status": "for production"
  }
]

Parse AI Output (Idea, Environment, Sound)

[
  {
    "Caption": "Blacksmith Forges Glowing Heart 🔨 #blacksmith #metalwork #forge #anvil #love #instagood #photooftheday #tbt #handmade #artistry #craftsmanship #viral",
    "Idea": "Blacksmith forges glowing metal heart on rustic anvil at dawn",
    "Environment": "Rustic forge interior, glowing embers, rough stone walls, cinematic close-up on anvil",
    "Sound": "Rhythmic hammer strikes and sizzling molten metal",
    "Status": "for production"
  }
]

AI Agent : Generate Detailed Video Prompts

Prompt (User Message)

Give me 3 video prompts based on the previous idea

Use the Think tool to review your output