Comfy-agent generate_sd15_crop_then_img2img

name: generate_sd15_crop_then_img2img description: Crop an input image and run SD1.5 img2img on the crop. inputs: image: type: string default: crop_source.png prompt: type: string default: refined editorial portrait based on the cropped source image negative_prompt: type: string default: watermark, text, blurry, low quality, distorted face x: type: integer default: 64 y: type: integer default: 64 width: type: integer default: 512 height: type: integer default: 512 steps: type: integer default: 24 denoise: type: float default: 0.5 seed: type: integer default: 73111924500188 outputs: image: string requirements: models: [] custom_nodes: [] links: [] input_modalities:

• text_prompt output_modalities:
• application/json model_families:
• other node_count: null node_types: [] selection_metadata: family: img2img_inpaint_outpaint resource_profile: unknown complexity_score: null estimated_runtime: depends on runtime parameters and server resources warnings:
• Hand-authored skill (no embedded workflow.json); metadata is inferred and may be approximate. max_width: null max_height: null max_steps: null