n8n: Generate video prompts with GPT-4o-mini

Step-by-step

1. Add an AI Agent node to the n8n workflow
2. Define the User Prompt by mapping product, features, ideal customer, and video settings from previous node inputs
3. Set the System Message to 'Expression' and input detailed instructions defining the agent's persona and output structure
4. Connect an OpenAI Chat Model sub-node to the AI Agent
5. Configure OpenAI credentials using a secret API key from the OpenAI platform
6. Select 'gpt-4o-mini' from the model dropdown list
7. Execute the node to generate the cinematic prompt

Options

• Use any available OpenAI model (e.g., base models or mini models)
• Define custom input fields for the prompt using colons and drag-and-drop mapping

Watch out for

• A chat model sub-node must be connected and enabled for the AI Agent to function
• OpenAI API requires a credit balance (e.g., $5) to generate keys and process requests
• Do not include the photo link in the prompt generation node as it is only needed for the final video API

Tips

• Use GPT-4o-mini models because they are cheaper than the base models
• Provide the AI with an example input and output format in the system message to ensure structured results
• Define the agent as an 'expert AI UGC video prompt engineer' to improve the quality of cinematic and emotional engagement