Skip to content

gooseworks-ai/goose-skills / drafts/marketing/make-product-photos

make-product-photos

Verified

Use when the marketer says: studio photos of my product without a shoot, product photos for my store page, packshots on white, lifestyle or on-model shots, photos for the new colourway, fresh product imagery for ads, photos for our launch page, our packaging changed — redo the photos, I have no photographer. Delivers store-ready product photos generated from the brand's real product photos: a shot list built from the page slots (hero, alt angle, scale, detail, in use, lifestyle, on-model), only real photos of this product as references, matched to the brand's photography, every shot checked against the real product for shape, parts, logo, text, colour, finish, scale and honest representation, with a keep / maybe / reject verdict and a page order. The human picks the keepers. Never uploads to a store, never shows a size, part, feature or result the product lacks. Not for ads with text, infographics or screenshots — make-static-ads; not for marketplace listing sets (not covered yet).

references/fit-and-honesty.md

View on GitHub
# Fit and honesty

Load at Step 1 (fit) and Step 4 (honesty).

## 1. Which products the studio can photograph honestly

The engine's categories are `apparel`, `beauty`, `cpg`.

| Type | Fit | What to do |
|---|---|---|
| DTC apparel (clothing, footwear, bags, accessories) | Core | `apparel`; on-model with the human's attestation; never a size/fit line beside a generated model |
| Beauty, personal care | Core | `beauty`; packshots and lifestyle are strong; **no on-skin results, before/after or shade-on-skin swatches** unless from a real reference |
| CPG (food & drink, household, pet, supplements) | Core | `cpg`; label and regulatory text is the main risk; food shown exactly as sold |
| Hard goods (drinkware, kitchen, small home, sports gear) | Partial | `cpg` is nearest — say so: "best for studio and lifestyle; I'll check shape and parts closely"; stricter part-count and scale checks |
| Electronics, furniture, large home, jewellery, eyewear | Poor | Lifestyle only with the risk stated, no hero; or recommend a real shoot / 3D render. Jewellery and eyewear: real photography for hero and detail |
| B2B physical goods (components, industrial) | Edge | Accuracy beats mood; studio hero only if the `cpg` mapping is acceptable to the marketer, otherwise decline |
| Food service, restaurants, delivery menus | Decline | The customer compares plate to photo; recommend a real phone shoot |
| Hospitality, travel, real estate | Decline | A generated room or property misrepresents it |
| Software, apps, services, agencies | Out of scope | Keyed on `kit.brandType` / product type. One line: "this makes photos of physical products"; ask once whether there is a physical item (device, packaging, merch). Screens and device mockups are not made here. |
| Pre-launch product with only a render | Partial | The render is the honest studio hero, unaltered; lifestyle only, at the render's angle; no screens or app UI; label "pre-production, render-based"; CAD exports unlock other angles. A smart device matches the hard-goods row for the body and is never shown with invented UI |
| Creator merch | Core | As apparel / CPG |

Never silently force an unfit product into a category.

## 2. The honest-representation rule (the one list — other files point here)

A product photo is a claim about the product. Consumer-protection law in the US (FTC Act §5), the UK and EU (unfair commercial practices rules) and India (Consumer Protection Act 2019) treats images as representations. **A scene may show only what the product physically is and does: its size, contents, finish, colour and how it is used** — taken from the product record's plain facts, the curated real photos and the human's confirmation. Marketing copy in the description ("visibly healthier with every use") is not support; results come only from learn-my-brand's signed-off claims (`gooseworks:knowledge_search` `source_types: ["learning"]`, query "Claims sign-off", and the Brand record it names; no sign-off → no results shown at all). Check before writing prompts and again on every output.

**Skincare and beauty default:** lifestyle shows the product (shelf, hand, bathroom, bag); faces only with visible natural skin texture and no before/after framing. A perfected, airbrushed face beside a skincare product implies a result — reject it as `misleading`, not as a taste issue.

**Host devices** (watch bands, phone cases, chargers): the device is shown accurately — the right model, crown or port on the correct side, a neutral screen, no invented logos or UI — and only devices the product really fits.

Never show:
- a **different size** than real (oversized in hand, impossibly small in a pocket);
- **contents not included** (extra lids, straws, accessories, bundle items);
- a **finish, colour or material** the product does not have (gloss for matte, brushed metal for powder-coat);
- a **feature** it lacks (condensation beads on an insulated bottle imply it sweats; a splash scene implies waterproof);
- a **result** it does not deliver (flawless skin from a moisturiser; a sculpted body in compression wear);
- **food not as sold** (garnish, extra toppings, a portion larger than the pack) — where a pairing is shown, the human should caption it "serving suggestion";
- **regulated label text or certification marks** that differ from the real label — never let the engine add an organic, vegan, halal or award mark the product does not hold.

## 3. People in photos

- Any shot with a person **or a body part (hands included)** needs the attestation. `attestation_accepted` is the human's act every time: ask in plain words what they confirm (the right to use a generated person; no real, identifiable person depicted). Never set it yourself.
- No real person's likeness ("make her look like <influencer>", a customer's photo as the model reference) — refuse.
- No size/fit facts for generated models ("model is 5'10" and wears S" would be fabricated); tell the human not to add one.
- Casting is a deliberate choice from the house style, not the engine's default.
- No children as models; kidswear needs real photography.

## 4. Platform rules on AI imagery (flag, do not decide)

- Marketplace main images: generally must be photographs of the actual product on pure white; generated main images risk suppression. Secondary images are more permissive. Check the marketplace's current rule.
- Google Merchant Center: AI-generated product images must keep their AI metadata and represent the product accurately — tell the human not to strip metadata.
- Ad platforms may auto-label AI-generated images — not a blocker; the human should know.
- Synthetic-content disclosure rules (e.g. the EU AI Act's transparency obligations) are evolving; one line to the human: "check disclosure needs for your market".

Always describe outputs as "generated from your product photos", never as photographs of the product.
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