media-forgemf-product-photolisted
Install: claude install-skill produtoramaxvision/maxvision-utilities
# media-forge:mf-product-photo
Ten named modes, a reference image, and a sentence of intent. Higgsfield's
backend assembles the structured prompt; the choice that matters is the mode.
## Intent
The prompt is not where the work is here. The backend already knows what a hero
banner looks like and what a moodboard pin looks like, and it writes a better
prompt for those than a generic description will. The work is picking the mode
that matches the deliverable, and looking at the assembled prompt before paying
for it.
## The ten modes
| Mode | Produces | Reach for it when |
|---|---|---|
| `product_shot` | Clean product on a controlled surface | Catalogue, PDP, the default |
| `hero_banner` | Wide, negative space for copy | Site header, email banner |
| `lifestyle_scene` | Product in a lived environment | Social, "in use" |
| `closeup_product_with_person` | Hands or partial figure with the product | Scale, tactility, texture |
| `virtual_model_tryout` | Product worn or held by a model | Apparel, accessories, wearables |
| `social_carousel` | A coherent multi-image set | Instagram carousel, a swipe sequence |
| `ad_creative_pack` | Several ad-shaped variations | Paid testing, a batch to choose from |
| `moodboard_pin` | Editorial, styled, Pinterest-shaped | Brand direction, aspirational |
| `conceptual_product` | Product rendered as an idea | Campaign, launch, abstract |
| `restyle` | Same product, different art direction | Refresh without a reshoot |
Mode drives aspect ratio