Style Me (Swag Mode)
Get AI-generated outfit recommendations based on a selfie, style preferences, and occasion. Each suggestion includes a virtual try-on image. Guardrails are enforced by Nano Banana 2.
Endpoint
Section titled “Endpoint”POST /style_meRequest body
Section titled “Request body”| Field | Type | Required | Description |
|---|---|---|---|
selfie_base64 | string | Yes | Base64-encoded selfie image |
style_preferences | string | No | Style keywords (e.g. “minimalist”, “streetwear”) |
occasion | string | No | Event or context (e.g. “casual work”, “date night”) |
num_outfits | number | No | Number of suggestions (1-4, default 3) |
store_image | boolean | No | If true, images are stored permanently. If false (default), images auto-delete after 24 hours. |
Example request
Section titled “Example request”curl -X POST https://mcll3bnfubyazfg6wekv3xc6fi0dgpyk.lambda-url.us-east-1.on.aws/style_me \ -H "X-API-Key: YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "selfie_base64": "BASE64_ENCODED_IMAGE", "style_preferences": "minimalist, clean lines", "occasion": "casual work", "num_outfits": 3 }'Response
Section titled “Response”| Field | Type | Description |
|---|---|---|
outfits | array | List of outfit suggestions |
Outfit object
Section titled “Outfit object”| Field | Type | Description |
|---|---|---|
description | string | Overall outfit description |
items | string[] | List of clothing pieces |
style | string | Style category |
image_url | string | Direct URL of the virtual try-on image. Auto-deletes after 24 hours when store_image=false (default), permanent when true. |
image_id | string | Unique image identifier. Use with GET /images/:id to retrieve metadata. |
mime_type | string | Image MIME type |
Example response
Section titled “Example response”{ "outfits": [ { "description": "Clean minimalist office look with neutral tones", "items": [ "Cream linen button-down shirt", "Charcoal tailored trousers", "White leather sneakers", "Silver minimalist watch" ], "style": "minimalist", "image_url": "https://superstyle-user-images.s3.us-east-1.amazonaws.com/temp/ss_abc123/style-me/20260228T153000Z_a1b2c3.jpg", "image_id": "img_a1b2c3d4e5f6a1b2c3d4e5f6", "mime_type": "image/jpeg" }, { "description": "Relaxed smart-casual ensemble", "items": [ "Navy cotton crew-neck sweater", "Khaki chinos", "Suede desert boots" ], "style": "smart casual", "image_url": "https://superstyle-user-images.s3.us-east-1.amazonaws.com/temp/ss_abc123/style-me/20260228T153000Z_d4e5f6.jpg", "image_id": "img_d4e5f6a1b2c3d4e5f6a1b2c3", "mime_type": "image/jpeg" } ]}- This is the most compute-intensive endpoint — expect 30-90 seconds per request
- Each outfit suggestion includes a generated try-on image
- If image generation fails for an outfit, the
image_urlfield will be empty - The AI analyzes body type, skin tone, and hair from the selfie to make appropriate suggestions