gemini-image-edit_skill

Overview

This skill edits and transforms existing images using text prompts powered by Gemini 3 Pro via fal.ai. It accepts one or multiple image URLs and returns edited images in common formats and resolutions. Use it to modify scenes, add or remove objects, change styles, or combine multiple references into a single cohesive result.

How this skill works

You provide a textual edit instruction plus one or more image URLs; the skill sends these to the Gemini 3 Pro image edit endpoint and returns generated image outputs and metadata. Optional parameters control number of outputs, aspect ratio, resolution, output format, and sync behavior. The service returns accessible image URLs and a short description of the edited result.

When to use it

• Modify an existing photo (color, lighting, season, or time of day)
• Add or remove objects or details from an image
• Apply a new visual style (watercolor, film, illustration)
• Combine elements from multiple reference images into one scene
• Generate alternative compositions or aspect ratios for a given image

Best practices

• Write specific, concise prompts describing exact changes and style cues
• Provide high-quality, publicly accessible image URLs to avoid 400 errors
• Start with lower resolution (1K) for faster iterations, then up-res for final outputs
• Use multiple reference images only when you want their elements combined or matched
• Enable sync_mode for quick data URI returns during prototyping

Example use cases

• Turn a summer landscape photo into a snowy winter scene for seasonal marketing
• Add furniture and decor to a real-estate photo to show staging options
• Convert product photos into consistent studio-style images for an ecommerce catalog
• Combine a model portrait and a background image to create a promotional banner
• Change lighting to golden hour for dramatic social media visuals