Canonical runtime-backed view of the artifact, independent from metadata governance.
{
"definition": {
"function": {
"description": "Generate an image or edit existing images based on a text prompt. The result will be a URI linking to the generated image. Use this tool when the user asks you to create a visual representation, mockup, or background image. Be highly descriptive in the prompt to get the best result.",
"name": "generate_image",
"parameters": {
"properties": {
"image_name": {
"description": "A short, snake_case filename to save the image as (e.g., 'cinematic_bg', 'hero_image').",
"type": "string"
},
"prompt": {
"description": "The detailed text prompt describing what to generate in the image.",
"type": "string"
}
},
"required": [
"prompt",
"image_name"
],
"type": "object"
}
},
"type": "function"
},
"kind": "tool",
"summary": {
"description": "Generate an image or edit existing images based on a text prompt. The result will be a URI linking to the generated image. Use this tool when the user asks you to create a visual representation, mockup, or background image. Be highly descriptive in the prompt to get the best result.",
"id": "generate_image",
"kind": "tool",
"path": "Tools/global/ai/generate_image.json",
"title": "generate_image"
},
"validation_issues": []
}