📦 Listing Manager - AI-Powered Multi-Platform Sales Tool

For non-technical users - Simple dashboard to manage products across eBay, Poshmark, and Mercari automatically.

---

Quick Start (5 Minutes)

Step 1: Install Dependencies

Open terminal and run:

cd ~/listing-manager
pip install -r requirements.txt

This installs all required packages (FastAPI for dashboard, requests for APIs, watchdog for folder monitoring).

---

Step 2: Start LM Studio

Before running the app, you need AI model access:

1. Download LM Studio (free)
2. Open LM Studio and load a vision model:
   • Click "Download Model"
   • Search for: llama3-llava-next-8b
   • Download this model (or similar llava-based vision model)
3. Go to "Server" tab (left sidebar)
4. Load the model you downloaded
5. Click "Start Server" (green button at bottom)
6. Make sure it says "Server running on http://localhost:1234/v1"

---

Step 3: Configure for Your System

Open config/settings.yaml and check these settings:

ai_parser:
  lmstudio_url: "http://localhost:1234/v1"  # This is correct if LM Studio on same machine
  
models:
  primary: "llava-next-nearest-34b"  # Use the exact model name you loaded in LM Studio
  fast: "qwen2.5-vl-7b-instruct"

platforms:
  ebay:
    enabled: true
    auto_relist_days: 30
    
  poshmark:
    enabled: true
    auto_relist_days: 30
    
  mercari:
    enabled: true
    auto_relist_days: 60

Important: The model name must match EXACTLY what you see in LM Studio's model list.

---

Step 4: Run the Dashboard

From terminal (in ~/listing-manager folder):

3. Start LM Server with model loaded: python main.py dashboard (or just python test_connectivity.py first)

Or just run normally first to test:

```bash
python main.py monitor

Then open your browser to: http://localhost:8000

---

How to Use Daily

Adding New Products (Simplest Way)

1. Create folder structure:

   ~/listing-manager/inventory/input/
   

2. Drop product photos into the input folder

3. The system automatically:

   • Scans for new images every few seconds
   • Analyzes each photo with AI (title, price, condition)
   • Flags items needing review
   • Moves processed items to processed/ folder

4. Check the dashboard at http://localhost:8000 to see results

---

Manual Single Product Scan

From terminal:

python main.py scan /path/to/product/photo.jpg

You'll see extracted details and can approve/reject before posting.

---

Features Explained

🤖 AI Product Analysis

• Extracts title, price, condition from photos
• Confidence score shows how certain the AI is (flag items below 75%)
• Uses your local LM Studio model - no cloud costs

🔄 Auto-Sync Across Platforms

When you sell an item on eBay:

1. System marks it as sold
2. Automatically removes listing from Poshmark & Mercari
3. Schedules relist after X days for fresh visibility boost

⏰ Smart Relisting

• eBay listings expire every 30 days → auto-relist with new ID
• Boosts search ranking without manual work
• Configurable per platform in settings.yaml

---

Troubleshooting

"AI analysis failed" errors

• Make sure LM Studio server is running (green status bar)
• Check model name matches exactly what's loaded in LM Studio
• Try a different model if one fails consistently

Platform connection issues

• The current version uses simulation mode (no actual API calls yet)
• To enable real posting: get OAuth/API credentials from each platform
• Update config/settings.yaml with your keys when ready

Dashboard won't start on port 8000

• Another service might be using that port
• Edit settings.yaml and change dashboard port to something else (e.g., 8080)

---

Moving to New Machine Later

This system is designed for easy migration:

1. Copy the entire ~/listing-manager/ folder
2. Install dependencies on new machine
3. Run LM Studio with same models
4. Update lmstudio_url in config if server IP changes
5. Done! All settings preserved

---

Support & Questions

• Check logs/app.log for detailed error messages
• Review config/settings.yaml to adjust timing, model choices
• Items below confidence threshold go to manual review queue

Happy selling! 🚀