comfyui

You can automate Stable Diffusion image generation by running API-format ComfyUI workflows with prompt, seed, image, and batch parameter injection. This handles the entire ComfyUI lifecycle locally or via Comfy Cloud, managing dependencies and capturing generated outputs.

An API-format workflow is a structured JSON representation of your ComfyUI node graph required for automated execution. This Skill uses these workflows to inject parameters directly, enabling text-to-image, upscaling, and animation pipelines without manual node wiring.

Yes, comfy-cli is a required dependency for lifecycle management when running ComfyUI media workflows. This relies on comfy-cli alongside websocket-client and torch to launch, validate, and route REST/WebSocket requests for local and cloud generation tasks.

Yes, you can run text-to-image and inpainting pipelines on Comfy Cloud. This supports cloud-aware model and file handling, allowing you to execute image, video, and audio generation workflows through either a local ComfyUI instance or the Comfy Cloud environment.

The best way to handle missing models in ComfyUI is to use an automation process that performs model checking and dependency validation before execution. This removes the friction of chasing missing models by validating dependencies and safely downloading outputs during workflow execution.

ComfyUI workflow execution may fail during batch generation if dependencies like torch or websocket-client are missing, or if the workflow is not in API-format. This prevents these issues by checking dependencies, validating models, and routing parameters correctly before execution.