Business Intelligence MCP Engine The Business Intelligence MCP Engine is a powerful all-in-one automation tool for: - Web scraping (basic & advanced) - Data extraction (auto / selector / text / images / headings) - Navigation & multi-step automation - Form detection and submission - Pagination crawling - Screenshot capture - SEO analysis - Change detection - Batch processing - Network/API monitoring - Link health checking This Actor provides a flexible MCP-style interface to interact with websites and extract structured data with minimal configuration. --- ## 🚀 Features - 25+ Operation Modes (Scraping, Extraction, SEO, Screenshots, Forms…) - Auto-detected extraction for simple pages - Full control with CSS selectors when needed - Headless browser automation using Playwright - Pagination crawling - Network request tracking - Batch URL execution - Change detection between HTML versions - Clean, metadata-rich output --- ## 🛠 Sample Input A minimal recommended input for new users: json { "operation": "scrape", "url": "https://example.com", "waitTime": 2000, "extractMode": "auto", "fullPage": true, "format": "png", "includeMetadata": true, "cleanContent": true } --- ## 🧰 All Supported Operations # Operation | Description -------------------------------------------------------------------------------- scrape | Basic scraping scrape_advanced | Deep scraping with browser automation extract_auto | Auto-detected extraction extract_selector | Extract using custom CSS selectors extract_images | Extract images extract_headings | Extract headings extract_text | Extract all visible text navigate | Multi-step navigation navigate_links | Crawl links from a page form_submit | Submit a simple form form_multi | Multi-field form submission form_detect | Auto-detect forms pagination_detect | Detect pagination pagination_crawl | Crawl paginated pages screenshot | Full-page or element screenshots screenshot_elements | Screenshots of multiple elements seo | SEO analysis changes | Detect changes between HTML versions article | Auto-extract article content batch | Batch scrape URLs batch_selectors | Batch scrape with selectors network | Capture all network requests network_api | Capture XHR/API requests only links_check | Check links for errors / redirects --- ## 📤 Output The Actor returns: * Cleaned HTML (optional) * Structured extraction data * Page metadata * List of network requests (if enabled) * Screenshots (if requested) * Diff/changes (for change detection mode) * Paginated or batch results in arrays --- ## ⚙️ Advanced Usage Extraction Using Custom Selectors json { "operation": "extract_selector", "url": "https://example.com/product", "selectors": { "title": { "selector": "h1", "type": "text" }, "price": { "selector": ".price", "type": "text" } } } --- ## 📝 Notes Some operations require url, others require html. Browser launches only when needed. The Actor handles anti-scraping protections using smart waits and retries. --- ## 📬 Support If you need custom automation, advanced scraping, or a private extension of this engine, feel free to contact the developer on Apify.