Company Contacts Finder

Apify actor (user vhsgreed) that turns a list of company names and/or website domains into DNS-verified contact emails.

How it works

1. For each input entry: entries containing a dot are treated as website domains; otherwise the official site is resolved with a DuckDuckGo HTML search (html.duckduckgo.com/html/?q=) — first external result host. No .com fallback guessing is used.
2. Fetches the homepage plus common contact paths: /contact, /contact-us, /about, /impressum, /kontakt (capped by maxPagesPerCompany, max 8).
3. Extracts emails from mailto: links and page text, including obfuscated forms like name(at)domain.com, name [at] domain [dot] com.
4. Verifies each unique email domain with an MX lookup via the Google DNS JSON API (https://dns.google/resolve?name=<domain>&type=MX). Emails whose domain has no MX record are dropped.
5. Flags freemail (gmail.com, hotmail.com, seznam.cz, …) and disposable (mailinator.com, 10minutemail.com, …) domains. Deduplicates per company.
6. Charges actor-start ($0.05) once per run and contact-found ($0.00125) per unique MX-valid email emitted. Max total charge $5/run.

Input

{
  "companies": ["alza.sk", "Mozilla Corporation"],
  "maxPagesPerCompany": 5
}

Output item

{
  "companyName": "Alza",
  "domain": "alza.cz",
  "emails": [
    {"email": "info@alza.cz", "mxValid": true, "isFreemail": false,
     "isDisposable": false, "sourcePageUrl": "https://www.alza.cz/kontakt"}
  ],
  "sourceUrls": ["https://www.alza.cz", "..."],
  "fetchedAt": "2026-09-25T12:00:00+00:00"
}

Verification (live sources, 2026-09-25)

• DuckDuckGo HTML search resolved alza.cz for "Alza official website contact" and correctly picked external hosts (no duckduckgo/ads links).
• Email extraction returned real emails from mozilla.org contact/about pages (trademark-permissions@mozilla.com, legal-notices@mozilla.com).
• MX lookups confirmed live records for mozilla.com/mozilla.org (Google Workspace MX) and correctly returned no MX for a nonexistent domain.
• Obfuscation parsing verified on synthetic HTML.

Known limitations (honest notes)

• Some large sites (e.g. alza.cz) block plain HTTP clients with 403/WAF (Cloudflare-style); no proxies or browsers are used, so emails from such sites cannot be extracted — the company is still emitted with an empty email list.
• Many sites render emails as images or via JS forms; nothing is found there.
• Site resolution from a company name is heuristic (first DDG external result) and can occasionally pick a different site than expected.
• Polite pacing: 0.5 s between requests, 20 s timeout per page.

Requirements

Python httpx + Apify SDK only. No browser, no proxies.