## Usage

### Full command (recommended):

The `--auto-fetch` flag will:
- Download the main product image from the product URL
- Search and download the brand logo
- Both get uploaded to ComfyDeploy automatically

## API Details

**Endpoint:** `https://api.comfydeploy.com/api/run/deployment/queue`
**Deployment ID:** `e37318e6-ef21-4aab-bc90-8fb29624cd15`

## ComfyDeploy Input Variables

These are the exact variable names the ComfyDeploy deployment expects:

| Variable | Type | Description |
|----------|------|-------------|
| `product_url` | string | Product page URL to scrape |
| `producto` | image URL | Product image (uploaded to ComfyDeploy) |
| `model` | image URL | Model/talent face reference |
| `referencia` | image URL | Style reference ad image |
| `marca` | image URL | Brand logo image |
| `brand_profile` | enum | Brand name from catalog |
| `prompt_profile` | enum | Funnel stage prompt |
| `aspect_ratio` | enum | Output format |

## 4-Phase Pipeline (How It Works Internally)

### Phase 1: Product Scraping
- Gemini Flash visits the product URL
- Extracts: title, description, features, price, images
- ⚠️ Image scraping is the most fragile part — always provide product images manually

### Phase 2: Campaign Brief Generation (CRITICAL)
- Uses Brand Identity JSON + Product Data → 10-point brief
- **Everything downstream depends on brief quality**
- Brief covers: strategic objective, central message, visual tone, product role, photographer, art direction, environment, textures, signature

### Phase 3: Blueprint Generation
- Master Prompt (per funnel stage) + Brief + Product JSON + Keyword Bank + Format
- Gemini Flash generates complete Blueprint JSON
- Covers: scene, production, graphic design, lighting, composition, materials, CTA

### Phase 4: Image Generation
- Nano Banana Pro (Imagen 3.0) generates the final image
- Uses Blueprint JSON + all reference images (product, talent, logo, style ref)

### Supporting Reference Nodes
- `pose_ref` → enforce a specific pose (replicated exactly)
- `photo_style_ref` → replicate photographic style (⚠️ can be too literal, being optimized)
- `location_ref` → replicate location and color palette

## Brand Profiles

### Existing catalog (70+ brands):

### Creating new brand profiles:
Use the `brand-analyzer` skill:

1. User provides: product URL + brand name + objective

2. CHECK brand profile exists:
   → ls ~/clawd/ad-ready/configs/Brands/ | grep -i "{brand}"
   → If not found: run brand-analyzer skill first
   
3. DOWNLOAD product image:
   → Visit the product URL in browser or fetch the page
   → Find and download the main product image
   → Save to /tmp/ad-ready-product.jpg

4. DOWNLOAD brand logo:
   → Search "{brand name} logo PNG" or fetch from brand website
   → Download clean logo image
   → Save to /tmp/ad-ready-logo.png

5. FIND reference image:
   → Search for "{brand name} advertisement" or similar
   → Or use a previously generated ad that has the right style
   → Save to /tmp/ad-ready-reference.jpg

6. SELECT prompt profile based on objective:
   → Awareness: brand discovery, first impressions
   → Interest: engagement, curiosity
   → Consideration: comparison, features
   → Evaluation: deep dive, decision support
   → Conversion: purchase intent, CTAs (most common)
   → Retention: re-engagement
   → Loyalty: brand advocates
   → Advocacy: referral, community

7. RUN the generation with ALL inputs filled

COMFY_DEPLOY_API_KEY="$KEY" uv run ~/.clawdbot/skills/ad-ready/scripts/generate.py \
  --product-url "https://shop.example.com/product" \
  --product-image "/tmp/product-photo.jpg" \
  --logo "/tmp/brand-logo.png" \
  --reference "/tmp/reference-ad.jpg" \
  --model "models-catalog/catalog/images/model_15.jpg" \
  --brand-profile "Nike" \
  --prompt-profile "Master_prompt_05_Conversion" \
  --aspect-ratio "4:5" \
  --output "ad-output.png"

COMFY_DEPLOY_API_KEY="$KEY" uv run ~/.clawdbot/skills/ad-ready/scripts/generate.py \
  --product-url "https://shop.example.com/product" \
  --brand-profile "Nike" \
  --prompt-profile "Master_prompt_05_Conversion" \
  --auto-fetch \
  --output "ad-output.png"