Cloud Integrations & Automations

📊

Google Sheets Automated Export

Automatically export newly extracted rows directly into Google Drive spreadsheets

● Connected & Ready

Connect your Google account to automatically sync scraped data into Google Sheets after every extraction run. Zero manual CSV uploading, zero middleware bottlenecks, and direct Drive synchronization.

⚡ Instant Auto-Export Generates or updates spreadsheets right when your scrape completes.

📑 Master Sheet Mode Optionally maintain a single master sheet that overwrites or appends runs.

🔒 Zero-Intermediary OAuth Direct browser-to-Google Drive sync without sending data to third parties.

🔐 Google Account Authorization

Your Google Account is linked to Urban Web Scraper. Extracted tables will sync automatically into your Google Drive root or designated spreadsheet folder.

Open Google Sheets Drive ↗

⚙️ Sync Preferences

Default Spreadsheet Title Pattern Available tokens: {name}, {date}, {time}, {rows}

Export Strategy Can also be switched per extraction in extension settings.

How to Use in Chrome Extension:

1

Open Extension Sidepanel

Click the Urban Web Scraper icon in Chrome to open your scraper dashboard.

2

Enable Google Sheets

Under Integrations, toggle Enable Google Sheets Export to ON.

3

Run Any Scrape

Every completed extraction immediately creates and populates your new spreadsheet.

🔗

Custom Webhooks Export

Receive instant HTTP POST notifications with extracted rows and CSV download links

HTTP POST Ready

Configure an HTTP or HTTPS endpoint to receive structured JSON notifications after every scraping task. Effortlessly connect to Zapier, Make.com, n8n, Pipedream, Airtable, Slack, Discord, or your internal API backend.

🔌 Webhook Endpoint Configuration

Destination Webhook URL Must be an active HTTP/HTTPS endpoint capable of accepting POST requests.

Signing Secret (HMAC-SHA256 Header: X-UrbanQode-Signature)

Use this key to cryptographically verify payload authenticity in your server.

Trigger Events

extraction.completed (Fires when scrape completes with extracted rows) extraction.failed (Fires if task encounters error or blocked target)

Standard JSON Payload Structure:

{
  "event": "extraction.completed",
  "task_id": "task_8fa12bc4",
  "task_name": "E-Commerce Catalog Extraction",
  "target_url": "https://example.com/products",
  "rows_extracted": 48,
  "duration_ms": 2850,
  "data": [
    { "title": "Wireless Studio Earbuds", "price": "$59.00", "rating": "4.8" },
    { "title": "Mechanical RGB Keyboard", "price": "$129.00", "rating": "4.9" }
  ],
  "download_csv_url": "https://api.urbanqode.com/api/tasks/task_8fa12bc4/export.csv",
  "timestamp": "2026-11-02T12:00:00.000Z"
}

🤖

Connect Claude (Desktop & Code)

Ask Claude to scrape web pages, extract tables into Markdown, and run autonomous crawls

Open Claude.ai ↗

Connect Claude Desktop or Claude Code to Urban Web Scraper using Model Context Protocol (MCP). Claude will inspect target pages, strip navigation & ads, parse tables into structured data, and run multi-page scrapes directly from your conversation.

⚡ One-Click AI Setup Prompt for Claude:

Copy this prompt and paste it into Claude Desktop or Claude Code. Claude will automatically configure its MCP server settings:

Option A: Claude Desktop Configuration

Add this server block to your claude_desktop_config.json:

{
  "mcpServers": {
    "urban-web-scraper": {
      "command": "npx",
      "args": ["-y", "urban-scraper-mcp"],
      "env": {
        "URBAN_API_TOKEN": "USER_TOKEN_HERE"
      }
    }
  }
}

Option B: Claude Code CLI Command

Run this single command in your terminal:

claude mcp add urban-web-scraper -- npx -y urban-scraper-mcp

Option C: Remote Cloud SSE Transport

If using a cloud or remote agent, enter the live SSE endpoint:

https://api.urbanqode.com/mcp/sse

What you can ask Claude:

"Scrape product prices" Extracts product names, variants, pricing, and stock into clean Markdown tables.

"Extract table to CSV" Pulls repeating tabular rows across paginated directories and formats CSV export.

"Find company contact details" Discovers emails, phone numbers, and social links from company about & team pages.

💬

Connect ChatGPT (Custom Actions & GPTs)

Equip Custom GPTs with web scraping and structured extraction capabilities

Open ChatGPT ↗

Connect OpenAI ChatGPT directly to the Urban Web Scraper engine via custom Actions. ChatGPT can crawl public web pages, extract tabular rows, analyze pricing, and format structured summaries on command.

How to Connect in ChatGPT:

1

Open GPT Editor

Go to ChatGPT GPT Editor and create or edit your custom GPT.

2

Add Custom Action

Under Actions, click Create new action and paste the OpenAPI schema below.

3

Set Authentication

Select API Key (Bearer) and provide your UrbanQode Access Token.

ChatGPT Custom Action OpenAPI 3.1 Schema:

{
  "openapi": "3.1.0",
  "info": {
    "title": "Urban Web Scraper Action API",
    "version": "1.0.0",
    "description": "Scrapes and extracts clean tabular data from web pages."
  },
  "servers": [
    { "url": "https://api.urbanqode.com" }
  ],
  "paths": {
    "/api/mcp/scrape": {
      "post": {
        "operationId": "scrapeWebpage",
        "summary": "Scrape public webpage to markdown",
        "parameters": [],
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "type": "object",
                "properties": {
                  "url": { "type": "string", "description": "Target webpage URL" }
                },
                "required": ["url"]
              }
            }
          }
        },
        "responses": {
          "200": { "description": "Clean structured markdown" }
        }
      }
    }
  }
}

Custom GPT System Instructions:

You are an elite data extraction agent powered by Urban Web Scraper.
When the user asks you to extract data from any URL:
1. Call the scrapeWebpage action with the target URL.
2. Parse the returned markdown into clean structured tables.
3. Provide summary metrics (total rows, key columns, price trends).
4. Offer download links or CSV formatting on request.

⚡

Connect MCP (Cursor, Windsurf & Agents)

Model Context Protocol support for IDEs, autonomous coders, and local agents

Full MCP Guide ↗

Model Context Protocol (MCP) is the universal open standard connecting AI coding assistants (Cursor, Windsurf, Claude Code, Antigravity) to local and cloud scraping tools.

Cursor IDE Setup (`.cursor/mcp.json`)

Paste this block into your project's .cursor/mcp.json or global Cursor Settings → Features → MCP:

{
  "mcpServers": {
    "urban-web-scraper": {
      "command": "npx",
      "args": ["-y", "urban-scraper-mcp"]
    }
  }
}

Remote Cloud Endpoint (SSE Transport)

For cloud agents, enter this URL directly:

https://api.urbanqode.com/mcp/sse

Exposed Autonomous Scraper Tools:

urban_scrape_page_markdown Strips ads/nav and converts target pages to token-efficient Markdown.

urban_extract_list Extracts repeating list containers and rows matching visual selectors.

urban_extract_shopify Extracts all products, variants, SKUs, and inventory from Shopify stores.

urban_extract_contacts Discovers emails, phone numbers, and social handles across website pages.

Connect Your Tools & AI Agents