Image Edit
Route everyday image edits—background swap, object removal, headline rewrite, batch i2i—through the best RunComfy edit model and its prompting patterns via one skill.
Overview
Image Edit is an agent skill most often used in Build (also Launch distribution, Validate prototype) that routes still-image edit requests on RunComfy to Nano Banana, GPT Image 2, Flux Kontext, or Z-Image Inpaint via the
Install
npx skills add https://github.com/runcomfy-com/skills --skill image-editWhat is this skill?
- Smart router across Nano Banana Edit (batch up to 20, identity-preserving default), GPT Image 2 Edit, Flux Kontext Pro,
- Bundles per-model prompting patterns to reduce wrong-model iteration
- Covers swap background, remove object, multilingual in-image text rewrite, multi-ref composition, and mask-driven region
- Calls `runcomfy run <vendor>/<model>/edit` through the local RunComfy CLI
- Triggers on image edit, image-to-image, i2i, and explicit single-or-batch edit requests
- Nano Banana Edit supports batch up to 20 images
- 4 edit models in the router: Nano Banana Edit, GPT Image 2 Edit, Flux Kontext Pro, Z-Image Turbo Inpaint
Adoption & trust: 113k installs on skills.sh; 1 GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You know what you want changed in an image but waste runs because each RunComfy edit model expects different prompts and strengths.
Who is it for?
Indie builders batching marketing variants or doing precise local edits who want the agent to match intent to RunComfy’s four documented edit routes.
Skip if: Pure canvas extension without in-frame editing focus (prefer image-outpainting), or teams not using RunComfy’s CLI-backed edit APIs.
When should I use this skill?
User asks for image edit, edit image, image-to-image, i2i, swap background, remove object, rewrite headline, or any explicit edit of one or many images.
What do I get? / Deliverables
The agent picks a catalog edit model, applies bundled prompting patterns, and returns sharper edits from `runcomfy run` without manual model shopping.
- Edited image file(s) from the selected RunComfy edit endpoint
- Model route aligned to batch size, mask need, or in-image text requirements
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Primary shelf is Build → integrations because the skill’s center of gravity is selecting and invoking `runcomfy run <vendor>/<model>/edit` for editing workflows. Image Edit is an agent-facing media API router, not application UI work—integrations is the correct canonical subphase.
Where it fits
Clean up a screenshot mock by removing clutter before user tests.
Batch up to twenty catalog shots with Nano Banana Edit for a consistent storefront look.
Rewrite in-image promo copy with GPT Image 2 Edit for multiple locales.
Mask-only logo placement on UGC stills using Z-Image Turbo Inpaint.
How it compares
Skill-pack router over four RunComfy edit endpoints—not a general-purpose desktop editor or a single-model one-shot prompt.
Common Questions / FAQ
Who is image-edit for?
Solo builders and content-focused founders using agent coding tools with RunComfy who need dependable image-to-image edits across backgrounds, objects, text-in-image, and masked regions.
When should I use image-edit?
In Build for asset production; in Validate for quick prototype visuals; at Launch for ad and storefront creative tweaks—whenever the user says image edit, i2i, swap background, remove object, or rewrite headline on a still.
Is image-edit safe to install?
Check this page’s Security Audits panel before trusting the package; edits run through RunComfy and may incur API cost, so scope CLI and network permissions deliberately.
SKILL.md
READMESKILL.md - Image Edit
# Image Edit — Pro Pack on RunComfy [runcomfy.com](https://www.runcomfy.com/?utm_source=skills.sh&utm_medium=skill&utm_campaign=image-edit) · [Nano Banana Edit](https://www.runcomfy.com/models/google/nano-banana-2/edit?utm_source=skills.sh&utm_medium=skill&utm_campaign=image-edit) · [GPT Image 2 Edit](https://www.runcomfy.com/models/openai/gpt-image-2/edit?utm_source=skills.sh&utm_medium=skill&utm_campaign=image-edit) · [Flux Kontext](https://www.runcomfy.com/models/blackforestlabs/flux-1-kontext-pro/image-to-image?utm_source=skills.sh&utm_medium=skill&utm_campaign=image-edit) · [Z-Image Inpaint](https://www.runcomfy.com/models/tongyi-mai/z-image/turbo/inpainting?utm_source=skills.sh&utm_medium=skill&utm_campaign=image-edit) · [GitHub](https://github.com/agentspace-so/runcomfy-skills/tree/main/image-edit) **Image edit, intent-routed.** This skill doesn't lock you to one model — it picks the right edit model in the RunComfy catalog based on what the user actually wants: batch identity-preservation, multilingual text rewrite, single-shot precise edit, or mask-driven region replacement. ```bash npx skills add agentspace-so/runcomfy-skills --skill image-edit -g ``` ## Pick the right model for the user's intent | User intent | Model | Why | |---|---|---| | Batch edit 1–20 images consistently (SKU gallery, A/B variants) | **Nano Banana Edit** | Up to 20 input images per call; locked aspect/resolution for series | | Swap background, preserve subject identity | **Nano Banana Edit** | Strong identity preservation under "keep X unchanged" prompts | | Localized object removal / addition with spatial language ("the left object", "upper-right corner") | **Nano Banana Edit** | Honors directional spatial scope | | Multilingual / non-Latin in-image text rewrite (Japanese kana, Cyrillic, Arabic) | **GPT Image 2 Edit** | Strongest in class for multilingual typography | | Multi-reference composition (subject from img1, scene from img2, palette from img3) | **GPT Image 2 Edit** | Numbered refs route cues correctly | | Layout-precise repositioning ("move headline from top-right to bottom-center") | **GPT Image 2 Edit** | Directional language honored at layout level | | Identity preservation across translated headline variants | **GPT Image 2 Edit** | Same source asset → many language variants, identity stable | | Single-shot precise local edit ("she's now holding an orange umbrella") | **Flux Kontext Pro** | Single-ref single-instruction, high-fidelity preservation | | Mask-driven object removal (cables, watermarks, distractions) | **Z-Image Turbo Inpaint** | Mask-required, strength-tunable, edge-consistent | | Mask-driven region replacement (full background swap with mask) | **Z-Image Turbo Inpaint** | High strength + clean mask = clean replacement | | Default if unspecified | **Nano Banana Edit** | Most flexible, supports both single and batch | The agent reads this table, classifies the user's intent, and picks the matching subsection below. ## Prerequisites 1. **RunComfy CLI** — `npm