Two approaches to images
Stock photos
The agent searches a library of free, high-quality stock photos from professional photographers. This works well when you need:- Real photos of people, places, or things
- Professional business imagery
- Lifestyle and authentic photos
- Generic backgrounds or hero images
Custom visuals
When stock photos don’t have what you need, the agent creates custom images using AI. This works well for:- Specific scenarios that don’t exist in stock libraries
- Unusual combinations or niche concepts
- Illustrations or stylized visuals
- Branded graphics or product mockups
- Anything too specific for stock photos
How the agent chooses
You don’t need to tell the agent which approach to use. It decides based on what you’re asking for. The agent uses stock photos when real, authentic images make sense. It creates custom visuals when stock photos won’t have what’s needed. Sometimes the agent tries both—searching stock photos first, and if nothing fits, creating a custom image.Creating custom images
The agent can generate entirely new images from text descriptions.How it works
You describe what you want, and the agent creates it. Be specific about:- What should be in the image
- The style (illustration, realistic, flat design, etc.)
- The mood or feeling
- Colors or visual direction
- “A minimalist illustration of a coffee cup in warm tones”
- “A modern, abstract background with blue and purple gradients”
- “A flat design icon showing a calendar and clock”
What you get
The agent creates the image and uses it directly in your app. You’ll see the image appear in the chat, and it’s automatically added to your project. Images are saved to your media library, so you can reuse them later if needed.Editing images
The agent can modify existing images based on your instructions.What can be edited
Change colors or elements”Change the red car to blue” “Make the sky look sunny instead of cloudy” Adjust the background
”Replace the background with a beach scene” “Make the background lighter” Modify parts of the image
”Add a person to the empty chair” “Remove the text overlay”
Removing backgrounds
The agent can remove backgrounds from any image to create transparent PNG files. This is useful for:- Creating cutouts for overlays
- Making product photos work on any background
- Preparing images for layered designs
- Creating logo or icon variations
Multiple images
The agent can work with multiple images at once when building pages that need several visuals. Example: “Build a features section with three images showing different benefits” The agent finds or creates three appropriate images and adds them to the section.Image quality and sizing
The agent handles technical details like image size, format, and optimization. Images are sized appropriately for where they’re used—large for hero sections, smaller for thumbnails, and optimized for web performance. You don’t need to specify dimensions or formats. The agent makes those decisions based on how the image will be used.How this helps you
The agent handling images means you can focus on what you want to build, not on finding or creating visuals. No design tools neededYou don’t need Photoshop, Figma, or image editing software. Describe what you want and the agent creates it. No stock photo browsing
The agent searches and evaluates options for you, picking images that fit your project. Faster iteration
Want to try a different image or style? Just ask. The agent can create variations or find alternatives quickly. You get professional-looking visuals without leaving the conversation or learning design tools.