generate_image

Generates an AI image for a brand using Nano Banana (Gemini). The prompt can be plain text OR a JSON object. When using JSON, keys are merged with the brand's default image prompt (set in BrandKit visual style), with per-image keys taking precedence. Example JSON prompt: {"motif": "Cute robots working in a garden", "aspect": "16:9"}. Returns the public URL of the generated image.

Available from 1 provider (1 remote)

Input Schema

            {
  "type": "object",
  "properties": {
    "brandId": {
      "description": "Brand ID (ObjectId)",
      "type": "string"
    },
    "prompt": {
      "description": "Image prompt — plain text or a JSON object. JSON keys merge with the brand's default image prompt.",
      "type": "string"
    },
    "model": {
      "description": "Model: 'gemini-3.1-flash-image-preview' (default, faster) or 'gemini-3-pro-image-preview' (higher quality)",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "aspectRatio": {
      "description": "Aspect ratio: '1:1', '3:4', '4:3', '9:16', '16:9', '21:9', '2:3', '3:2', '4:5', '5:4' etc. (default: model decides)",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "containsText": {
      "description": "Whether the image should contain rendered text (e.g. infographics). Default false.",
      "type": "boolean",
      "default": false
    }
  },
  "required": [
    "brandId",
    "prompt"
  ]
}