What problem does it solve? Choosing the right AI image model and writing correct API calls for text-to-image or image-to-image tasks is tedious and error-prone. This Skill routes each request to the appropriate model on RunComfy and builds the exact runcomfy run CLI invocation with the correct schema. ## Core Features & Use Cases - Model routing for text-to-image: Picks among FLUX 2 Klein, GPT Image 2, Nano Banana 2, Seedream 5, Qwen Image, Wan 2-7, and Z-Image Turbo based on intent such as typography, photoreal portraits, or sub-second iteration. - Image-to-image editing: Routes edits to Nano Banana 2 Edit, GPT Image 2 Edit, or FLUX Kontext Pro for identity-preserving edits, multilingual text swaps, and single-instruction local changes. - Documented prompt patterns: Ships per-model prompting tips, schema tables, and ready-to-run CLI commands. - Use Case: A marketer needs a multilingual poster with an exact Japanese headline. The Skill selects GPT Image 2, quotes the headline text precisely, and runs the command with the correct size parameter. ## Quick Start Ask the agent to generate an image of your subject, for example a photoreal portrait or a poster with an exact headline, and it will pick the right model and run the RunComfy CLI command.
