nano-banana-edit

Name: nano-banana-edit
Availability: InStock
Author: doany

Edit up to 20 images per call with identity-preserving spatial prompts.

5|2|Updated May 18, 2026

Search Tags:#workflow-automation #image-editing #identity-preservation #batch-editing #runcomfy #spatial-language #nano-banana-edit

Ready to run

npx skills add https://github.com/doany-ai/skills --skill nano-banana-edit-doany-ai

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nano-banana-edit
Source: https://github.com/doany-ai/skills/tree/main/nano-banana-edit
Command: npx skills add https://github.com/doany-ai/skills --skill nano-banana-edit-doany-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Edit images with Google Nano Banana 2 Edit endpoint to preserve subject identity, swap backgrounds, and localize edits with spatial language, supporting batch edits up to 20 inputs.

Core Features & Use Cases

Identity-preserving edits across multiple images
Batch processing of up to 20 images in a single call
Spatially localized edits and selective references via prompts
Route to GPT Image 2 edit / Flux Kontext / Nano Banana 2 t2i as needed

Quick Start

Invoke the RunComfy nano banana edit workflow with your image URLs and a precise prompt.

Frequently Asked Questions about nano-banana-edit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I perform identity-preserving batch image edits for multiple photos?▼

Identity-preserving batch image edits apply subject consistency across multiple photos by submitting up to 20 image URLs in a single API call. The Nano Banana Edit endpoint processes the batch while maintaining the original subject identity.

Can I use spatial language to constrain localized edits in image processing?▼

Spatial language constrains localized edits by allowing prompts to target specific regions within an image. This enables selective modifications like background swaps or regional object changes without altering the untargeted areas.

What is the maximum number of images I can process in a single batch editing call?▼

The maximum number of images for batch editing is 20 per call. You provide the image URLs, prompt, and parameters like aspect ratio and resolution in the JSON body to process them simultaneously.

How do I automate image editing workflows using RunComfy?▼

Automate image editing workflows by invoking the RunComfy nano banana edit endpoint with a JSON body. Pass image URLs, prompt, and configuration parameters to trigger automated, identity-preserving edits.

Does the Nano Banana edit endpoint support routing to other models like GPT Image 2 or Flux Kontext?▼

The Nano Banana edit endpoint routes to GPT Image 2 edit, Flux Kontext, or Nano Banana 2 t2i as needed. This allows specialized handling for different image generation and editing scenarios.

What JSON parameters are required for spatially constrained image edits?▼

Required JSON parameters for spatially constrained edits include prompt, image_urls, and number_of_images. Optional fields like aspect_ratio, resolution, seed, and safety_tolerance further refine the output.