Route image editing requests to RunComfy edit models by intent.
npx skills add https://github.com/runcomfy-com/skills --skill image-edit-runcomfy-com
Please help me install this Agent Skill. Skill: image-edit Source: https://github.com/runcomfy-com/skills/tree/main/image-edit Command: npx skills add https://github.com/runcomfy-com/skills --skill image-edit-runcomfy-com
It solves the problem of getting consistent, high-quality image edits without manually choosing the best generative model for the user’s exact intent.
Use image-edit to replace a headline in a poster image while preserving the photo, layout, and brand mark.
High-intent search queries and answers about installing and using this skill.
Intent-based model routing automatically directs image editing requests to the appropriate RunComfy edit model based on user intent, handling batch edits, multilingual text rewrites, local edits, and mask-driven inpainting.
Batch processing supports 1–20 image edits with locked aspect and resolution to maintain a consistent series, while preserving identity during background swaps or headline rewrites across the entire gallery.
Mask-driven inpainting enables precise object removal or background region replacement by supplying a grayscale mask with tunable strength, executing spatial instructions via the RunComfy CLI.
Multilingual text rewrites support replacing non-Latin in-image typography, allowing you to update headlines or text overlays in multiple languages while preserving the original photo and layout.
Model invocation requires RunComfy CLI authentication through the runcomfy login command or by setting the RUNCOMFY_TOKEN environment variable, ensuring validated input JSON fields are accepted per selected route.
Single-ref high-fidelity local edits are ideal for single-image precision adjustments, whereas batch processing handles 1–20 images simultaneously for consistent series updates like product SKU galleries.
runcomfy-comInpaint masked image regions via RunComfy CLI with JSON input.
GroundMountCompanyConvert approved product images into cinematic video variations with multiple AI models.
prime-skillsRoute image editing requests to the appropriate RunComfy model and execute them.
RheasilviaGenerate music tracks from lyrics and descriptive tags using a local foundation model.
