/create_image

name: Create Image
allowed-tools: Bash, mcp__replicate__create_models_predictions, mcp__replicate__get_predictions
description: Generate image(s) via Replicate

Create Image

This command generates image(s) based on the provided prompt using Replicate.

Variables

DROPPED_FILE_PATH: [[FILE_PATH]] DROPPED_FILE_PATH_ARCHIVE: agentic_drop_zone/generate_images_zone/drop_zone_file_archive/ IMAGE_OUTPUT_DIR: agentic_drop_zone/generate_images_zone/image_output/<date_time>/

• This is the directory where all images will be saved
• The date_time is the current date and time in the format YYYY-MM-DD_HH-MM-SS

MODEL: google/nano-banana ASPECT_RATIO: 16:9

Workflow

• First, read `DROPPED_FILE_PATH`.
• Create output directory: `IMAGE_OUTPUT_DIR/<date_time>/`
• IMPORTANT: For every image prompt detailed in the `DROPPED_FILE_PATH` do the following:

<image-loop>

• Use `mcp__replicate__create_models_predictions` with the MODEL specified above
• Pass image prompt as the prompt input
• Use ASPECT_RATIO for the image dimensions
• Wait for completion by polling `mcp__replicate__get_predictions`
• Save the executed prompts to `IMAGE_OUTPUT_DIR/<date_time>/image_prompt_<concise_name_based_on_prompt>.txt`
• Use the exact prompt that was executed in the `mcp__replicate__create_models_predictions` call
• Download the image: `IMAGE_OUTPUT_DIR/<date_time>/<MODEL_NAME_underscore_separated>_<concise_name_based_on_prompt>.jpg`
• Display:
• Generation time
• File size
• Full path to saved image

</image-loop>

• After all images are generated, open the output directory: `open IMAGE_OUTPUT_DIR/<date_time>/`
• When you finish generating images, move the `DROPPED_FILE_PATH` into a `DROPPED_FILE_PATH_ARCHIVE` directory