Скачать или смотреть What's new in Bolt v2

X: https://x.com/boltdotnew
YouTube:    / @boltdotnew  
Join our discord: https://discord.bolt.new
Jakub’s twitter: https://x.com/jakub_works
Help Center: https://support.bolt.new/
V2 launch announcement: https://x.com/boltdotnew/status/19730...

Prompt in video:
Context:
Build me an AI virtual try-on web app where users can see themselves wearing different clothes. The user uploads a picture of themselves, uploads a few clothing items (hat, top, pants, bag, accessory, extra) and our app generates the person with those clothing items on them. We should also add a text field that users can use instead of adding specific items, so they can describe the changes they want to make to the picture. Save the generated images in Bolt Database. Users should see only their own generated images.
Requirements:
CRITICAL: Make sure the prompt tells the AI to keep the person's face and background identical but change only their clothes. There should be no cropping the image ever, always 1:1 with the original.
Technical considerations:
For the backend, use Bolt Database.
Use gemini-2.5-flash-image API for image generation, for which I’ll provide the API key later.
Please add it as a secret and use Bolt Database edge functions.
Design:
Keep the design simple and professional with neutral colors - no bright colors. Everything should be monochromatic grays, blacks, and whites.