Skip to content

beatra-ai/beatra-skills / plugins/product-photo-studio/skills/product-photo-studio

product-photo-studio

Verified

Transform a real product photo into a studio-quality ecommerce image, lifestyle scene, or marketplace-ready hero shot. This AI product photography tool replaces backgrounds, improves lighting, and stages scenes while using the source photo and confirmed product details as the visual anchor. Create clean white-background listings, contextual lifestyle compositions, and premium ad visuals from a single phone snap for Amazon, Taobao, Shopify, and social media. Start from one product photo, combine several references, or refine a selected draft toward a polished listing image.

references/scene-craft.md

View on GitHub
# Scene craft

## Anchor the result to the product

The product is the subject. Build the prompt around the source photo and
confirmed product details, then verify fidelity in the result:

- state explicitly that Image 1 is the product reference and ask the model to
  match its shape, color, label, and proportions closely;
- describe only the background, lighting, and scene around it;
- do not request changes to the product itself in a transform call—use
  `beatra.images.edit` for localized product fixes.

## Background quality

A professional product background has three properties:

- **clean edges** — the product's outline must have no halo, fringe, or
  ghosting where it meets the background;
- **consistent lighting** — the light direction on the product must match the
  light direction implied by the background (shadows fall the same way);
- **natural shadow** — the product should cast a subtle contact shadow on the
  surface beneath it; a floating product with no shadow looks artificial.

Include these constraints in the prompt: "clean product edges, consistent
lighting direction, natural contact shadow, no halo or fringe."

## Clean white background

For marketplace main images requiring pure white:

- state "pure white background, RGB 255 255 255" explicitly;
- request even, shadowless lighting for the background while keeping a subtle
  contact shadow under the product for grounding;
- avoid gradients, textures, or patterns unless the user explicitly requests
  them.

## Lifestyle scene composition

For lifestyle and contextual scenes:

- match the surface material to the product category (wood for food, marble
  for cosmetics, metal for tech);
- keep the product as the focal point—larger and more prominent than any prop
  or background element;
- use depth of field to separate the product from the background (sharp
  product, slightly soft background);
- ensure color temperature consistency (warm scene = warm product lighting,
  cool scene = cool product lighting).

## Lighting direction

State the light direction in the prompt to avoid conflicting shadows:

- **front light** — even, flat illumination; good for catalog accuracy;
- **side light** — dimensional, shows texture; good for food and crafts;
- **backlight** — dramatic silhouette rim; good for glass and transparent
  products;
- **top light** — natural overhead; good for flat-lay and top-down shots.

The chosen direction must be consistent between the product lighting and the
scene's implied light source.

## Color accuracy

Use the original photo as the color reference, then review the result for color
fidelity:

- do not describe the product's color in the prompt (let the reference image
  determine it);
- if the original photo has a color cast, note it and advise the user that
  color correction may be needed;
- structured `palette` is optional—use it only when the user provides exact
  brand or background color values.

## Canvas and output

- Default canvas for marketplace main images: square `1:1` at `2K` tier.
- For banner or hero images: `16:9` or `4:3` at `2K`.
- For edit: the default is `source`, which inherits the base image's ratio.
- A different ratio is a separate paid request with its own confirmation and
  `client_request_id`.
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 196kUpdated Claude CodeCodexVerifiedInstall