Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Product photography enhancement for ad creatives using AI image generation. Takes a product image and generates 5 professional photography styles for ad use: Studio, Floating, Ingredient, In Use, and Lifestyle. Requires GOOGLE_API_KEY or configured ADS_IMAGE_PROVIDER. Triggers on: "product photo", "product photography", "photoshoot", "enhance product image", "product shoot", "product photos for ads", "studio shot", "lifestyle photo"
.claude/skills/miosa-osa-ads-photoshoot/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 5% | 0% |
> Transforms a product image or description into professional ad-ready photography > in 5 distinct visual styles, each at platform-appropriate sizes.
bash/ads photoshoot /ads photoshoot --styles studio floating /ads photoshoot --product shoe.jpg /ads photoshoot --all-platforms
| Flag | Type | Default | Description | |------|------|---------|-------------| | --styles | string] | all 5 | Which styles to generate | | --product | string | — | Product image file path | | --all-platforms | flag | false | Generate all 5 sizes per style |
bashexport GOOGLE_API_KEY="your-key" # See /ads generate for alternative providers
Check for brand-profile.json — if found, inject:
colors.primary into backgrounds and accentsaesthetic.mood_keywords as atmosphere descriptorstarget_audience for Lifestyle and In Use contextimagery.forbidden exclusions into all promptsClean, e-commerce product shot. White seamless background, even studio lighting.
Dramatic levitation effect with gradient background.
Flat lay with components/materials artfully arranged. Top-down view.
Authentic usage context. Hands using product, shallow depth of field. Hands only — no full face (avoids model release complications).
Aspirational full-context shot. Environmental context with product as hero.
./product-photos/
studio/
product-studio-1080x1080.png
product-studio-1080x1920.png
floating/
product-floating-1080x1080.png
product-floating-1080x1920.png
ingredient/
product-ingredient-1080x1080.png
in-use/
product-in-use-1080x1080.png
product-in-use-1080x1920.png
lifestyle/
product-lifestyle-1080x1080.png
product-lifestyle-1080x1920.png| Style | Best Platforms | Rationale | |-------|---------------|-----------| | Studio | Meta Feed, LinkedIn, Google PMax | Universal, clean, platform-safe | | Floating | Meta Reels, TikTok, Stories | High visual impact on vertical | | Ingredient | Meta Feed, Pinterest | Best as square; tells product story | | In Use | TikTok, Meta Reels, Stories | Authentic, native-feeling content | | Lifestyle | All platforms | Aspirational, broad audience appeal |
brand-profile.json — from /ads dna (optional)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 17,449 | 17,828 | +2% | 1 | 1 | 0% | 2,741 | 3,700 | +35% | 0 | 0 | — |
case-02 | fail→fail | 13,130 | 22,303 | +70% | 1 | 1 | 0% | 1,625 | 4,445 | +174% | 0 | 0 | — |
case-03 | fail→fail | 18,341 | 26,538 | +45% | 1 | 1 | 0% | 3,235 | 5,466 | +69% | 0 | 0 | — |
case-04 | pass→pass | 16,130 | 12,955 | -20% | 1 | 1 | 0% | 2,123 | 3,297 | +55% | 0 | 0 | — |
case-05 | pass→pass | 7,487 | 4,940 | -34% | 1 | 1 | 0% | 1,089 | 2,058 | +89% | 0 | 0 | — |
case-06 | pass→pass | 19,108 | 22,673 | +19% | 1 | 1 | 0% | 3,252 | 4,435 | +36% | 0 | 0 | — |
case-07 | fail→pass | 10,376 | 19,908 | +92% | 1 | 1 | 0% | 1,800 | 2,874 | +60% | 0 | 0 | — |
case-08 | fail→fail | 15,983 | 9,839 | -38% | 1 | 1 | 0% | 2,148 | 2,374 | +11% | 0 | 0 | — |
case-09 | fail→fail | 7,942 | 11,951 | +50% | 1 | 1 | 0% | 1,187 | 2,922 | +146% | 0 | 0 | — |
case-10 | pass→pass | 10,588 | 10,456 | -1% | 1 | 1 | 0% | 1,575 | 2,551 | +62% | 0 | 0 | — |
case-11 | pass→pass | 15,202 | 11,556 | -24% | 1 | 1 | 0% | 2,069 | 2,744 | +33% | 0 | 0 | — |
case-12 | pass→pass | 9,341 | 7,976 | -15% | 1 | 1 | 0% | 1,351 | 2,079 | +54% | 0 | 0 | — |
case-13 | pass→pass | 26,177 | 12,855 | -51% | 1 | 1 | 0% | 2,378 | 2,794 | +17% | 0 | 0 | — |
case-14 | pass→pass | 15,920 | 10,698 | -33% | 1 | 1 | 0% | 2,415 | 2,574 | +7% | 0 | 0 | — |
case-15 | pass→pass | 15,145 | 10,449 | -31% | 1 | 1 | 0% | 2,487 | 2,373 | -5% | 0 | 0 | — |
case-16 | pass→pass | 12,900 | 8,610 | -33% | 1 | 1 | 0% | 2,088 | 2,303 | +10% | 0 | 0 | — |
case-17 | fail→pass | 8,535 | 3,375 | -60% | 1 | 1 | 0% | 1,514 | 1,528 | +1% | 0 | 0 | — |
case-18 | fail→pass | 8,228 | 2,969 | -64% | 1 | 1 | 0% | 1,370 | 1,497 | +9% | 0 | 0 | — |
case-19 | pass→pass | 10,872 | 3,513 | -68% | 1 | 1 | 0% | 1,487 | 1,450 | -2% | 0 | 0 | — |
case-20 | fail→pass | 10,780 | 4,209 | -61% | 1 | 1 | 0% | 1,673 | 1,753 | +5% | 0 | 0 | — |
case-21 | fail→pass | 12,059 | 4,407 | -63% | 1 | 1 | 0% | 1,522 | 1,783 | +17% | 0 | 0 | — |
case-22 | fail→pass | 14,486 | 4,266 | -71% | 1 | 1 | 0% | 2,198 | 1,819 | -17% | 0 | 0 | — |
case-23 | fail→pass | 13,341 | 9,524 | -29% | 1 | 1 | 0% | 2,089 | 2,212 | +6% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +35 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.