Compatibility record
Does Replit Agent support Image input?
Replit Agent accepts image and screenshot inputs on Starter, Core, Teams, and Enterprise so users can attach visual references or ask it to refine a design from a screenshot.
What this state means
Supported, with the qualification kept attached.
Users can attach images and screenshots in the prompt box. Agent can also screenshot a pasted URL. Visual context is for understanding references, not native image generation.
Accepts images and interprets their visual content as part of a request.
Image attachment, OCR, screenshot inspection, and video understanding are not interchangeable.
Researcher note
Path A: chat upload guide (images/screenshots) + updates changelog (attach images to prompts with Agent). Learn context-management independently lists screenshots and images as model context. Docs do not exclude Starter.
Evidence
Primary records behind the answer.
- 01
“From the prompt box, select the plus button, then choose Upload a file. You can also drag and drop a file into the prompt box.”
- 02
“As always, you can always attach images to your prompts or mention files with @file.”
- 03
“Screenshots, images, or design references”