skills.rest
runcomfy-com
runcomfy-comOFFICIAL
Explore all skills by this author →
Website𝕏@runcomfyaiGitHub

gpt-image-edit

Edit images via OpenAI GPT Image 2 endpoint using RunComfy CLI.

12|2|Updated May 18, 2026
Search Tags:#image editing#multi-reference#ad creatives#identity preservation#runcomfy cli#gpt image 2#multilingual text
One-click install
npx skills add https://github.com/runcomfy-com/skills --skill gpt-image-edit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gpt-image-edit
Source: https://github.com/runcomfy-com/skills/tree/main/gpt-image-edit
Command: npx skills add https://github.com/runcomfy-com/skills --skill gpt-image-edit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

You need high-quality image edits that preserve the subject’s identity and brand elements while reliably changing things like backgrounds, layout, and in-image text—without manual rework.

Core Features & Use Cases

  • Identity-preserving edits: Keeps faces, poses, clothing, framing, and brand marks unchanged while applying targeted changes.
  • Multilingual in-image text rewriting: Replaces text inside the image in many scripts (Latin, kana, CJK, Cyrillic, Arabic) with strong control via quoting the exact characters.
  • Single- and multi-reference composition: Uses up to 10 images as reference inputs for scene/layout matching and composition cues.

Quick Start

Ask the skill to keep the person and brand mark unchanged while replacing only the headline text in the image.

Frequently Asked Questions about gpt-image-edit

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

FAQPage Schema
How do I edit image backgrounds while preserving the subject's identity and brand marks?▼

Identity-preserving image edits keep faces, poses, and brand marks unchanged while applying targeted background replacements. You provide a prompt and image URLs to execute the modification safely.

Can I rewrite multilingual text inside an image without breaking the original layout?▼

Multilingual in-image text rewriting replaces text across Latin, kana, CJK, Cyrillic, and Arabic scripts with strong control. You quote the exact characters in the prompt to maintain layout precision.

How many reference images can I use for multi-reference composition in image editing?▼

Multi-reference composition supports up to 10 HTTPS image URLs as reference inputs. You provide these images to guide scene matching and layout cues for the final generated output.

What image sizes are supported when performing identity-preserving image-to-image edits?▼

Identity-preserving edits support size values of auto, 1024x1024, 1024x1536, and 1536x1024. You specify the desired dimensions within the JSON input before invoking the runcomfy CLI.

Why use GPT Image 2 for ad creatives instead of other image editing tools in the same category?▼

GPT Image 2 applies identity-preserving edits specifically for brand-safe ad creatives. It reliably changes backgrounds and in-image text without manual rework, distinguishing it from other category-level tools.

Do I need a specific CLI environment to execute GPT Image 2 edit endpoints?▼

Executing GPT Image 2 edits requires the RunComfy CLI environment. You pass a JSON input containing the prompt and image URLs to invoke the openai/gpt-image-2/edit endpoint and download outputs.

Related Skills

Explore More Design & Creative Skills→
agentspace-soagentspace-so
31

gpt-image-2

Generate and edit brand-safe images with GPT Image 2 on the RunComfy Model API.

Official
Intermediate
akrindevakrindev

youtube-downloader

Download YouTube videos and playlists with audio, subtitles, and metadata.

Community
Intermediate
prime-skillsprime-skills
31

image-edit

Route image editing requests to the appropriate RunComfy model and execute them.

Official
Advanced
ovachieverovachiever
52

reportlab

Generate PDF documents with ReportLab's Canvas and Platypus APIs.

Community
Advanced