• DATA
  • MONITORING
  • INVENTORY
  • HUMAN

How to extract product data with your AI

Connect Diffbot to your AI client once. From there, your agent handles the heavy lifting of turning raw HTML into clean, usable data.

Ask AI about this page

Short answer

How can I pull product info from websites using AI?

Your AI uses Diffbot to scan retail pages and pull specific details like prices and SKUs into a structured format. You get clean JSON data ready for your spreadsheets or databases.

The results.

Where your data lands.

The AI processes the raw web content and delivers specific outputs based on your request.

DATA

Structured JSON files

The AI converts messy website layouts into organized JSON objects. You get clean fields for price, SKU, and name.

MONITORING

Price tracking logs

By extracting price data repeatedly, your AI builds a history of cost changes. This makes it easy to spot trends.

INVENTORY

SKU and stock lists

The agent pulls specific product identifiers and availability status. You can feed this directly into your own inventory tools.

HUMAN

Manual verification

If a site has weird layouts or bot protections, the AI will flag it. You decide when to manually check a specific URL.

The workflow

What your AI does when a URL arrives.

The AI moves from raw page inspection to structured data extraction.

  1. Scans the page

    The AI looks at the target URL to understand the layout and identify product elements.

    analyze_page

  2. Pulls product details

    The agent identifies specific attributes like price, name, and SKU from the page content.

    extract_product

  3. Formats the output

    The AI organizes the found information into a standard JSON structure.

    extract_product

  4. Validates data

    The agent checks that the extracted fields match the schema you requested.

    analyze_page

Try it

Copy these to start.

Use these prompts to trigger the extraction process.

Starting points

Swap out the bracketed info for your actual target URLs or specific field names.

Diffbot Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_FiT28RroNiKkMcafLWUuAoeV2pbdaAk9JBpgecYE/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Diffbot capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "diffbot-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_FiT28RroNiKkMcafLWUuAoeV2pbdaAk9JBpgecYE/mcp"
    }
  }
}
Copy into chat04

  • Extract the price and SKU from this product page: [URL]

  • Use Diffbot to get a JSON list of all product names and prices on this page: [URL]

  • Analyze this URL and tell me if the product is currently in stock: [URL]

  • Pull the product description and manufacturer info from this link: [URL]

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Diffbot once, then ask.

One click connects the MCP to your client. Your credentials stay encrypted, and you can find the full technical details on the Diffbot Connector page.

Connect Diffbot to your AI

FAQ

How this task behaves.

  • 01

    Can the AI change the price on the website?

    No. The AI only reads and extracts data from the page; it cannot modify any external websites.

  • 02

    Will it work on sites that require a login?

    No. The current tools are designed to analyze public-facing product pages.

  • 03

    Can I get the data in a CSV instead of JSON?

    The tools output structured JSON, but you can ask your AI client to convert that JSON into a CSV format for you.

  • 04

    Does the AI handle image extraction?

    Yes, the extraction process can pull image URLs as part of the product data.

  • 05

    What happens if the website layout changes?

    The AI uses the page analysis tool to re-identify elements, but if a site is heavily obfuscated, you may need to manually verify the results.

  • More questions about Diffbot? The Connector page answers them. See everything the Diffbot Connector can do

Connect Diffbot to Claude, Cursor, ChatGPT & more