Web Intelligence Suite (v2.2)

High-speed technical web auditor and extractor: W3C WCAG 2.1/2.2 Level AA, European Accessibility Act (EAA 2025), German BFSG, US ADA Title III, OWASP Security Headers, AI/RAG Markdown Converter, and Universal E-Commerce Product Intelligence.

---

Overview

Web Intelligence Suite is a unified, serverless cloud engine designed for digital agencies, compliance officers, data engineers, and AI agent developers. It combines comprehensive technical accessibility auditing, LLM-ready content extraction, security analysis, and e-commerce scraping into a single high-performance pipeline.

Unlike traditional crawlers that spin up resource-heavy headless browsers (Puppeteer/Playwright) consuming 1–2 GB of RAM per run, Web Intelligence Suite uses an optimized DOM analysis pipeline powered by Cheerio and Axios. It processes pages in 200–500 milliseconds using only 128 MB of memory, reducing execution costs and latency by over 80%.

---

Key Capabilities

1. Web Accessibility Audit (WCAG 2.1/2.2 AA, EAA 2025, BFSG, ADA)

Automated technical validation aligned with W3C WCAG 2.1/2.2 Level A/AA criteria:

• Form Controls & Labels (WCAG 3.3.2 / 4.1.2): Identifies inputs, selects, and textareas missing programmatic <label>, aria-label, or aria-labelledby.
• Text Alternatives (WCAG 1.1.1): Detects images missing alt attributes or with invalid empty text.
• Language Declaration (WCAG 3.1.1): Validates presence of valid lang attribute on the root <html> element.
• Heading Hierarchy (WCAG 1.3.1): Verifies presence of main <h1> landmarks and logical heading progressions.
• Viewport Zoom Restrictions (WCAG 1.4.4): Detects mobile user-scalable=no and maximum-scale=1.0 viewport locks.
• CSS Selectors & Code Snippets: Every finding includes exact DOM CSS selectors and code snippets for instant developer remediation.

2. Third-Party Resource & Privacy Hygiene (GDPR / DSGVO)

• External CDN Fonts: Identifies unhosted external font dependencies (e.g. fonts.googleapis.com, Adobe Typekit) that pose GDPR compliance liabilities in the European Union.
• Third-Party Trackers & Pixels: Audits presence of external analytics and advertising scripts (Google Analytics, Meta Pixel, TikTok, Hotjar).
• Consent Management Platform (CMP): Automatically detects installed CMP banners (Usercentrics, Cookiebot, OneTrust, Borlabs Cookie, Complianz, Axeptio).

3. OWASP HTTP Security Headers & Metadata

• Security Headers: Audits essential OWASP headers: HTTP Strict Transport Security (HSTS), Content Security Policy (CSP), X-Frame-Options, X-Content-Type-Options, and Referrer-Policy.
• SEO & Social Graph: Validates <title>, meta descriptions, canonical URLs, and OpenGraph/Twitter Card metadata.
• Schema.org Structured Data: Extracts top-level JSON-LD schema types (Organization, LocalBusiness, FAQPage, Product, Article).

4. AI / RAG Clean Markdown Extractor

• Strips navigation bars, footers, scripts, styles, and promotional noise.
• Converts main content, article bodies, headings, lists, and HTML tables into clean, structured Markdown.
• Calculates exact estimated token counts for context window optimization in ChatGPT, Claude, and LangChain agents.

5. Universal E-Commerce Product Intelligence

• Automatically extracts structured product data across Shopify, WooCommerce, Shopware, Magento, and Schema.org Product markup.
• Normalizes product name, SKU, GTIN/EAN, price, currency, availability, and customer review metrics.

---

Executive Visual HTML Reports

When auditing websites, the Actor automatically compiles an interactive, executive Visual HTML Audit Report and stores it in the run's default Key-Value Store:

• Single Page Scan: Detailed breakdown with compliance score (0–100), letter grade (A–F), metric cards, and remediation code cards.
• Multi-Page Site Crawl: Aggregated domain health score, page-by-page audit table, and prioritized critical violation inventory.
• Print to PDF: Built-in CSS print styling allows digital agencies to export white-label PDF audit reports for clients directly from their browser.
• Access: Available under the Key-Value Store tab as OUTPUT_REPORT.html.

---

Operation Modes

Mode	Identifier	Description	Use Case
Full Suite	full	Complete WCAG audit, security check, Markdown extraction, and e-commerce scraping.	Default comprehensive domain analysis.
Accessibility Only	accessibility	High-speed accessibility and compliance audit only (WCAG, EAA, BFSG, ADA).	Fast client compliance audits & reporting.
AI / RAG Markdown	markdown	Clean Markdown extraction and token estimation without audit scoring.	Feeding live web content into LLMs & RAG pipelines.
E-Commerce Only	ecommerce	Product price, stock, SKU, and schema extraction.	Price monitoring and catalog synchronization.

---

Input Configuration

{
  "startUrls": [
    { "url": "https://example.com" }
  ],
  "mode": "full",
  "crawlInternalLinks": true,
  "maxPages": 20,
  "generateVisualReport": true
}

Parameter Reference

• startUrls (array, required): One or more URLs to process. You can provide a homepage, multiple links, or a sitemap URL.
• mode (string, optional, default: "full"): Pipeline mode ("full", "accessibility", "markdown", or "ecommerce").
• crawlInternalLinks (boolean, optional, default: true): When enabled, discovers and audits internal same-domain links automatically.
• maxPages (integer, optional, default: 20): Total ceiling on audited pages per run to protect budget limits.
• generateVisualReport (boolean, optional, default: true): Generates OUTPUT_REPORT.html in Key-Value Store for visual inspection.

---

Output Data Structure

Dataset Overview Columns (Table View)

Field	Type	Description
finalUrl	String	Audited page URL after redirects.
domain	String	Extracted target domain.
score	Integer	Overall compliance and quality score (0–100).
grade	String	Letter rating (A, B, C, F).
criticalIssues	Integer	Number of critical accessibility or security violations.
tokens	Integer	Estimated token count of clean Markdown content.
durationMs	Integer	HTTP response and parsing latency in milliseconds.
statusCode	Integer	HTTP response status code (e.g. 200).

---

Programmatic Integration Examples

Python (apify-client)

from apify_client import ApifyClient

# Initialize the client with your Apify API token
client = ApifyClient("YOUR_APIFY_TOKEN")

# Run the Actor
run_input = {
    "startUrls": [{"url": "https://example.com"}],
    "mode": "accessibility",
    "crawlInternalLinks": True,
    "maxPages": 10
}

run = client.actor("lightoton/web-intelligence-suite").call(run_input=run_input)

# Fetch results from the output dataset
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(f"URL: {item.get('finalUrl')} | Score: {item.get('score')} | Grade: {item.get('grade')}")

JavaScript / Node.js (apify-client)

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });

const run = await client.actor('lightoton/web-intelligence-suite').call({
  startUrls: [{ url: 'https://example.com' }],
  mode: 'markdown',
  maxPages: 5,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
for (const item of items) {
  console.log(`Title: ${item.title} | Tokens: ${item.tokens}`);
}

LangChain / AI Agent Tool

from langchain.tools import Tool
from apify_client import ApifyClient

apify_client = ApifyClient("YOUR_APIFY_TOKEN")

def extract_web_markdown(url: str) -> str:
    run = apify_client.actor("lightoton/web-intelligence-suite").call(
        run_input={"startUrls": [{"url": url}], "mode": "markdown", "crawlInternalLinks": False, "maxPages": 1}
    )
    items = list(apify_client.dataset(run["defaultDatasetId"]).iterate_items())
    return items[0].get("markdown", "") if items else "Failed to extract content."

web_extractor_tool = Tool(
    name="WebToMarkdownExtractor",
    func=extract_web_markdown,
    description="Extracts clean, noise-free Markdown with tables and headers from any target URL."
)

---

Pricing

Monetized via Apify Pay-Per-Event:

• $10.00 / 1,000 pages processed ($0.01 per audited URL).
• No flat monthly fees. Pay only for pages successfully processed.