product-description-generator

**Required:**

• Platform (eBay / Walmart / Shopify / Etsy / TikTok Shop / Lazada / Shopee)
• Product name and key features
• Brand name

**Recommended (better results):**

• 1-3 competitor URLs to analyze
• Target audience
• Tone preference (Professional / Friendly / Urgent / Luxury)

Which mode?

• **A — Create**: I'm writing a new listing from scratch
• **B — Optimize**: I have an existing listing to improve

💡 For Amazon listings, I recommend using amazon-listing-optimization — it's optimized for Amazon's Cosmo algorithm.

---

## Mode A Workflow — Create New Listing

### Step 1: Collect Product Info

| Field | Required | Example |
|-------|----------|---------|
| `product_name` | ✅ | Portable blender |
| `platform` | ✅ | Etsy |
| `brand` | ✅ | BlendJet |
| `key_features` | ✅ | USB-C, 6 blades, BPA-free |
| `specs` | ✅ | 380ml, 175W motor |
| `target_audience` | 👍 | Gym-goers, travelers |
| `use_cases` | 👍 | Smoothies, protein shakes |
| `competitor_urls` | 👍 | 1-3 URLs to analyze |
| `tone` | Optional | Professional (default) / Friendly / Luxury / Urgent |

### Step 2: Gather Keywords

**If competitor URLs provided:**

1. Fetch each competitor page:

Use web_fetch on each competitor URL.

Extract: title, bullets/features, description, price, review count, brand name.

2. If `web_fetch` fails or returns incomplete data:

Fallback: web_search for "[product title from URL]" site:[platform].com

Extract data from search snippets.

3. Parse competitor content and extract keywords in these categories:
   - **Product-type terms**: What it IS (yoga mat, exercise mat)
   - **Feature terms**: What it DOES (non-slip, eco-friendly)
   - **Use-case terms**: WHERE/WHEN used (home gym, yoga studio)
   - **Audience terms**: WHO buys (beginners, athletes)
   - **Attribute terms**: Specs (6mm, TPE material)

4. Expand beyond competitors:

web_search: "[product type]" best seller features what buyers want

web_search: "[product type]" review complaints common issues

web_search: site:[platform].com "[product type]"

**If no competitor URLs provided:**

1. Discover keywords via web search:

web_search: "[product name]" best seller [platform] features

web_search: "[product name]" review what customers love

web_search: "[product name]" vs alternatives comparison

web_search: site:[platform].com "[product name]"

2. Extract keywords from top 5 results following the same categories above.

⚠️ **Critical**: Remove all competitor brand names — never include them in output.

### Step 3: Score and Prioritize Keywords

Score each keyword (1-9 points):

| Dimension | Scoring |
|-----------|---------|
| **Frequency** | In 3+ competitor titles = 3 pts / In 1-2 = 2 pts / Bullets only = 1 pt |
| **Relevance** | Core descriptor = 3 pts / Feature = 2 pts / Peripheral = 1 pt |
| **Opportunity** | Few competitors use = 3 pts / Most use = 2 pts / All use = 1 pt |

Assign to tiers:

🔴 Primary (7-9 pts) → Title

🟡 Secondary (4-6 pts) → Bullets / Features

🟢 Tertiary (2-3 pts) → Description

⚪ Backend (1 pt) → Tags / Search Terms

### Step 4: Generate Copy

Proceed to **Generate Copy** section.

---

## Mode B Workflow — Optimize Existing Listing

### Step 1: Analyze Current Listing

User may provide:
- **Full listing copy** (title, bullets, description pasted directly)
- **Product URL** (e.g., `https://www.etsy.com/listing/123456`)
- **Platform + product identifier** (e.g., "Etsy listing 123456")

**If user provides URL or identifier only:**

Use web_fetch on the provided URL.

Extract: current title, bullets/features, description, tags (if visible), price.

If `web_fetch` fails:

Fallback: web_search for the product title or identifier.

Ask user to paste the listing content manually if data is incomplete.

**Once listing content is obtained**, parse and extract:
- All keywords currently present
- Structure and format used
- Obvious gaps (missing features, weak benefits, no FABE structure)

### Step 2: Gather Target Keywords

**If competitor URLs provided:**

Follow the same competitor analysis process as Mode A Step 2:
1. `web_fetch` each competitor URL
2. Extract their keywords
3. Expand via web search

**If no competitor URLs provided:**

Discover ideal keywords for the product type:

web_search: "[product type]" top keywords [platform] 2024 2025

web_search: "[product type]" best seller features

web_search: site:[platform].com "[product type]" top listings

### Step 3: Gap Analysis

Compare current keywords vs. target keywords:

Keyword Gap Analysis

### ✅ Keywords You Already Have

| Keyword | Title | Bullets | Description |

|---------|-------|---------|-------------|

| yoga mat | ✅ | ✅ | ✅ |

| exercise mat | ❌ | ✅ | ❌ |

### ❌ Keywords You're Missing

| Keyword | Priority | Recommendation |

|---------|----------|----------------|

| non-slip | 🔴 High | Add to title |

| eco-friendly | 🟡 Medium | Add to bullet 2 |

| extra thick | 🟡 Medium | Add to bullet 3 |

Current Coverage: 12/20 keywords (60%)

Target Coverage: 90%+

### Step 4: Rewrite

Generate optimized copy incorporating missing keywords.

Show **Before → After** for each component.

Proceed to **Generate Copy** section.

---

## Generate Copy

Final step for all modes after keyword priority table is built.

### Writing Framework: FABE

Apply to every bullet:

F — Feature: What the product HAS or DOES

A — Advantage: Why this is BETTER than alternatives

B — Benefi