๐กWant ad-ready images without designers?
๐This workflow automatically takes an image, analyzes it, and generates a professional ad-ready version using AI. Letโs walk through how it works. ๐ Step 1 โ Form Trigger โWe start with a simple form trigger. Here, I can upload an image and give it a short description. This kicks off the entire workflow.โ ๐ Step 2 โ Upload to Google Drive โNext, the workflow uploads that image to Google Drive. This makes the file accessible for the next AI steps.โ ๐ Step 3 โ AI Image Analysis โThen, the image is analyzed using OpenAIโs vision model. It checks whether the picture is mainly a product, a character, or both, and describes the visual details in structured YAML format.โ ๐ Step 4 โ Prompt Generation โBased on that analysis, another AI agent creates a short and natural image prompt. This prompt is carefully designed for ad use, making sure the product looks authentic, the text stays accurate, and the style feels realisticโlike a genuine ad photo.โ ๐ Step 5 โ Structured Parsing โTo keep things consistent, the prompt is passed through a structured parser. This ensures the final text is clean and ready to use.โ ๐ Step 6 โ Image Generation via OpenRouter โNow comes the fun part! The workflow sends both the original image and the AI-generated prompt to the OpenRouter API, which calls Googleโs Gemini 2.5 Flash model. This generates a brand new ad-style image.โ ๐ Step 7 โ File Conversion & Save โThe generated image data is processed, converted into a file, and stored. So at the end, I get a ready-to-use ad image that matches my original product shot, but looks polished for marketing.โ ๐drop โad-readyโ in the comments to get the file