home/ecommerce/cars-full

Cars Full

GPTClaudeGemini··965 copies·updated 2026-07-14
cars-full.prompt
You are scraping a car dealership inventory website.

Step 1 - Inventory pages:
Go to the inventory page. You will find car listings displayed as cards.
Each card shows the car's basic info and a link to its detail page.
Only process the first page. Do NOT follow any pagination links.
Only extract the first 2 cars from the page.

For each car, extract the data in this exact JSON format:
{
  "Year": 2020,
  "Make": "Toyota",
  "Model": "4Runner",
  "Price": "$34,995",
  "detail_url": "https://example.com/Inventory/Details/abc123-def456"
}

IMPORTANT: The detail_url must be the EXACT href from the <a> tag linking to that car's
detail page. Look for the anchor tag wrapping each car card or the "View Details" /
"More Info" link. Copy the href value exactly as-is — do NOT construct, guess, or
modify URLs. If the href is relative (e.g. /Inventory/Details/...), convert it to
an absolute URL using the current page's domain.

Put all pagination links in next_urls.
Put each car's detail page URL in detail_urls.

Step 2 - Detail pages:
Visit each car's detail page. The detail page has:
- A photo gallery/carousel with multiple car images. Look for all image URLs
  in ![alt](url) markdown patterns, <img> tags, or any URLs ending in
  .jpg, .jpeg, .png, .webp. Collect ALL image URLs, not just the first one.
- A specs/details section with VIN, transmission, colors, drivetrain, fuel type,
  mileage, engine, stock number, condition, and other vehicle information.
- Dealer info: dealership name, address, phone number, and zip code.

Extract all available data from the detail page in this JSON format:
{
  "Year": 2020,
  "Make": "Toyota",
  "Model": "4Runner",
  "Price": "$34,995",
  "Condition": "Used",
  "Mileage": "45,000",
  "Engine": "4.0L V6",
  "Transmission": "Automatic",
  "Drivetrain": "4WD",
  "Fuel": "Gasoline",
  "Exterior_Color": "Midnight Black",
  "Interior_Color": "Black",
  "Stock#": "T12345",
  "VIN": "JTEBU5JR8L5123456",
  "Fuel_Economy_City": "16",
  "Fuel_Economy_Hwy": "19",
  "detail_url": "https://example.com/inventory/vehicle-detail-url",
  "images": ["https://example.com/img1.jpg", "https://example.com/img2.jpg", "...every image URL found on the page"],
  "Agent_Name": "Western Star Auto Sales",
  "Agent_Address": "1701 N. Cicero Ave, Chicago, IL, 60639",
  "Agent_Phone": "(773) 887-4766",
  "zip": "60639"
}

when to use it

Community prompt sourced from the open-source GitHub repo souvik79/Scraper-AI (MIT). A "Cars Full" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

ecommercecommunitygeneral

source

souvik79/Scraper-AI · MIT