Turn a product brief or URL into a UGC video ad

{
  "type": "object",
  "required": [
    "brief"
  ],
  "properties": {
    "brief": {
      "type": "text",
      "maxLength": 2000,
      "minLength": 10,
      "description": "What it is, who it's for, what to land. URL works."
    },
    "music": {
      "type": "string",
      "default": "auto",
      "maxLength": 120,
      "description": "Background music for the clip's native soundtrack. `auto` usually picks a quiet instrumental bed matching your brand's tone (lo-fi, soft acoustic, or light upbeat — always under the voice); raw-phone concepts like ASMR stay music-free. `none` always skips music for a raw phone-recording feel. Or type a style hint (e.g. \"lo-fi chill\", \"upbeat electronic\").\n"
    },
    "captions": {
      "type": "boolean",
      "default": true,
      "description": "Burn word-synced karaoke captions onto the finished clip — the retention look for muted TikTok/Reels/Shorts feeds. Caption colors and styling are picked automatically to match the product's brand identity. Turn off for a clean, caption-free clip.\n"
    },
    "last_frame": {
      "type": "image",
      "description": "Optional. A still the clip should end on (tail keyframe). When set, the render lands on this exact frame — e.g. an end card, or the last frame of a gameplay recording so the clip flows seamlessly into the next shot when stitched. Pairs with `first_frame` to interpolate from open to close.\n"
    },
    "first_frame": {
      "type": "image",
      "description": "Optional. A still the clip should open on (image-to-video). When set, the render starts from this exact frame — e.g. the first frame of a gameplay recording so the clip flows seamlessly into the real footage when stitched.\n"
    },
    "ugc_template": {
      "enum": [
        "auto",
        "unboxing",
        "tryout",
        "interview",
        "tutorial",
        "product_review",
        "problem_solution",
        "grwm",
        "asmr_demo",
        "pov",
        "before_after",
        "listicle",
        "reaction",
        "storytime"
      ],
      "type": "string",
      "default": "auto",
      "description": "Which UGC format to shoot. `auto` picks the best fit."
    },
    "aspect_ratios": {
      "type": "array",
      "items": {
        "enum": [
          "9:16",
          "1:1",
          "16:9"
        ],
        "type": "string"
      },
      "default": [
        "9:16"
      ],
      "minItems": 1,
      "description": "Output frame(s). Pick one or more — one clip is rendered per selected ratio.",
      "uniqueItems": true
    },
    "creator_image": {
      "type": "image",
      "description": "Optional photo of the on-camera person, for a consistent creator."
    },
    "product_images": {
      "type": "array",
      "items": {
        "type": "image"
      },
      "maxItems": 5,
      "minItems": 0,
      "description": "Optional. Product photos that anchor the real product. Sharp, well-lit shots work best. If omitted, the product is rendered from the brief/research description."
    },
    "creator_persona": {
      "type": "text",
      "maxLength": 300,
      "description": "Optional. Who's on camera and where. Empty = auto-cast."
    },
    "duration_seconds": {
      "type": "integer",
      "default": 8,
      "maximum": 15,
      "minimum": 4,
      "description": "Target clip length. 8 s is the sweet spot; up to 15 s supported. Longer clips buy more action, not more talking — spoken lines stay short either way, so pick longer durations for demos and reactions, not bigger scripts.\n"
    }
  }
}

{
  "type": "object",
  "properties": {
    "videos": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "video_url": {
            "type": "video",
            "description": "Drive path to the rendered UGC clip (with native audio); served to readers as a stable media URL. The playground renders it with a <video> player."
          },
          "aspect_ratio": {
            "enum": [
              "9:16",
              "1:1",
              "16:9"
            ],
            "type": "string",
            "description": "The frame this clip was rendered for."
          }
        }
      },
      "minItems": 1,
      "description": "One rendered clip per aspect ratio."
    }
  }
}