Website Contact Extractor (Browser): JS-Heavy Company Sites

Extract contacts (name, position, email, phone) from JavaScript-rendered company sites (React, Vue, Angular SPAs) using Playwright plus AI. Browser-based companion to the HTTP-only Website Contact Extractor: handles team, contact, about and impressum pages that return empty HTML without JavaScript execution. Same input format, same output, anti-hallucination validation against source HTML. Auto-chained when the HTTP actor flags a company with js_rendering_suspected.

When to use this actor

• JS-rendered team/contact pages: React, Vue, Angular and other SPA frameworks
• Empty HTML responses: sites where the contact page is rendered client-side
• Companies flagged by the HTTP actor: auto-chained via enablePlaywrightFallback
• Team detail pages: two-phase render captures individual member profiles
• Anti-hallucination requirement: contacts are validated against source HTML, not LLM-invented

How it works

• Link discovery from the homepage via HTTP (fast)
• Phase 1: Playwright renders main pages (team, contact, about, impressum)
• Phase 2: Playwright renders team member detail pages
• LLM extraction using Gemini Flash, Groq or OpenRouter
• Anti-hallucination validation against source HTML
• Cross-company dedup to catch LLM-invented names

API example

# Run the browser contact extractor on a list of JS-heavy company sites
curl -X POST "https://api.apify.com/v2/acts/santamaria-automations~website-contact-extractor-browser/runs?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "companies": [
      {
        "company_id": "abc-123",
        "company_name": "TechCorp AG",
        "website_url": "https://techcorp.ch"
      }
    ],
    "llmProvider": "gemini",
    "geminiApiKey": "YOUR_GEMINI_KEY"
  }'

# Or use with AI agents via MCP:
# https://mcp.apify.com?tools=santamaria-automations/website-contact-extractor-browser

Integrations

• n8n, Make, Zapier: trigger runs and process contact records via webhook
• AI Agents (MCP): Claude Desktop, Cursor, VS Code, LangChain, LlamaIndex
• Python, Node.js: Apify SDK for programmatic access
• CRM (HubSpot, Pipedrive, Salesforce): push contacts directly into your pipeline

Output sample

{
  "company_id": "abc-123",
  "company_name": "TechCorp AG",
  "contacts": [
    {
      "name": "Max Mustermann",
      "position": "CTO",
      "email": "[email protected]",
      "confidence": 0.92
    }
  ],
  "status": "success",
  "browser_extraction": true
}

Related Actors

• Website Contact Extractor (HTTP): lean HTTP-only version, run this first
• Website Job Extractor (Browser): same Playwright stack, job-extraction output
• URL Contact Scraper: single-fetch contact extraction at scale