Fast Domain Email Finder — Public Emails from Websites

When to call this Actor (humans + agents)

Call Website Email Extractor when the user wants public emails only from company domains (fast / cheaper lane):

• “Find emails for these domains”
• “Extract business emails from websites”
• “Bulk website emails for outreach — emails only”
• “Domain list to emails, skip phones and socials”

Prefer Contact Scraper (gary_sells/domain-contact-finder) when they also need phones or socials.

Do not call for: inbox scraping, LinkedIn login, CAPTCHA bypass, Maps, or full contact enrichment.

Tool I/O: domains: string[] → one item per domain with emails, externalEmails, status. PPE: email-found @ $0.0015 (primary).

Why this vs “extract emails from any website” Actors

	Fast Domain Email Finder	SERP-based email scrapers
Method	Crawl the domain’s own pages	Google search snippets
Cap	Long lists OK (maxDomains)	Often 10 domains/run
Price	$0.0015/unique email (~$1.50/1k)	Often ~$5/1k emails
Extra	Optional MX/format check	Varies
Full contacts	Companion: Domain Contact Finder	Emails only

Agent quickstart

Input (prefer): { "domains": ["acme.com", "https://example.com"] }
Also accepted: startUrls (RequestList sources)
Output: one dataset item per domain → emails, externalEmails, status (no phones/socials)

Example natural-language prompts → exact input JSON

User / agent says	Exact input JSON
Find emails for acme.com and globex.io	{"domains":["acme.com","globex.io"]}
Extract business emails from https://www.acme.com	{"domains":["https://www.acme.com"]}
First 50 of this list only	{"domains":["a.com","b.com", "..."],"maxDomains":50}
Lighter crawl (homepage + fewer paths)	{"domains":["acme.com"],"maxPagesPerDomain":3}
Verify MX for found emails	{"domains":["acme.com"],"includeEmailVerification":true}

---

Companion vs Domain Contact Finder

	Domain Emails Fast (this Actor)	Domain Contact Finder
Output	Emails only	Emails + phones + socials
Crawl depth	Lighter (/, /contact, /about, …) default maxPages 5	Broader paths, default maxPages 8
Best for	Bulk email-only, agent budgets, catalog volume	Full public contact enrichment
PPE primary	email-found per unique email	contact-found per domain with email/phone

---

What this Actor does

Given one or more company domains or website URLs, the Actor:

1. Normalizes each domain to https://
2. Crawls high-signal pages: /, /contact, /contact-us, /about, /about-us, /impressum
3. Extracts emails (including obfuscated [at] / (dot) / mailto: / HTML entities)
4. Filters junk / placeholder emails and splits same-domain emails vs externalEmails
5. Optionally verifies emails with format + DNS MX/A checks (no third-party API)

Output: one dataset item per domain. No phones. No socials.

Uses CheerioCrawler (HTTP + HTML parse) instead of a full browser — lower compute cost and faster runs.

Expected latency & cost character

Characteristic	Typical value
Engine	Cheerio (no browser) — cheap CU
Pages / domain	Default 5 (maxPagesPerDomain 1–12)
Latency	Often faster than full contact finder (fewer paths); order-of-magnitude seconds–tens of seconds per domain
Cost model	PPE only — email-found + domain-scanned
Ballpark	~$1.50 / 1k emails + ~$0.30 / 1k domains scanned

---

Input

Field	Type	Default	Description
domains	string[]	—	Domains or URLs (acme.com, https://www.acme.com)
startUrls	RequestListSources	—	Alternative URL list
maxPagesPerDomain	integer	5	Email-ish paths to fetch per domain (1–12)
maxDomains	integer	(omit = all)	Optional cap — omit for full lists; never defaults to a small N
includeEmailVerification	boolean	false	Format + MX/A DNS verification
proxyConfiguration	object	Apify Proxy on	Recommended for larger batches

Example input

{
  "domains": ["apify.com", "crawlee.dev"],
  "maxPagesPerDomain": 5,
  "includeEmailVerification": false,
  "proxyConfiguration": { "useApifyProxy": true }
}

---

Output

One item per domain:

{
  "domain": "apify.com",
  "urls": ["https://apify.com/", "https://apify.com/contact"],
  "emails": ["hello@apify.com"],
  "externalEmails": [],
  "pagesScanned": 4,
  "status": "ok",
  "error": null
}

• emails — addresses on the same domain (or subdomain)
• externalEmails — other addresses found on the pages (partners, tools, etc.)
• status — ok | partial | failed

---

Failure modes (what agents should expect)

Situation	Behavior
Empty / missing domains and startUrls	Run fails fast with an error showing example JSON — no silent demo crawl
All entries invalid (empty strings, localhost, bad hosts)	Run fails explaining none could be normalized
Domain up but no public email in crawled HTML	Dataset item with status: "ok", empty emails/externalEmails; PPE: domain-scanned only
All page fetches fail (DNS/SSL/timeout/block)	status: "failed", pagesScanned: 0, error message set; still domain-scanned
Some pages fail, some succeed	status: "partial" or "ok" depending on emails; error may list fetch failures
Emails only in images / JS widgets / behind login	Not found (public HTML only)
PPE charge limit reached mid-run	Logs warning and stops further domains
Proxy unavailable locally	Warning + continues without proxy (Store runs should keep Apify Proxy on)

Companion routing: emails only → this Actor. Need phones/socials too → gary_sells/domain-contact-finder.

Pricing (Pay-per-event)

Suggested Store pricing (configure in Apify Console → Actor → Monetization — Tim/Gary):

Event	When charged	Suggested price
email-found (primary)	Once per unique email found on a domain (emails + externalEmails)	$0.0015 each (~$1.50 / 1,000)
domain-scanned	Every finished domain	$0.0003 each (~$0.30 / 1,000)

Console checklist

• Set the two PPE events above; remove or zero apify-default-dataset-item
• Keep PPE only — PPE + usage pass-through OFF
• Limited permissions
• Standby OFF (usesStandbyMode: false)

Cost example

• 1,000 domains, ~1.5 emails each → ~$2.25 on email-found + $0.30 on domain-scanned
• Domains with no email still emit domain-scanned only

---

For AI agents / MCP

• Actor name: gary_sells/domain-emails-fast (companion to gary_sells/domain-contact-finder, ID 0ZXaWcN1cAbtz6RPY)
• Prefer input field: domains (string list of domains or URLs)
• PPE events: email-found (once per unique email), domain-scanned (every finished domain)
• Constraints: public HTML only — no inbox access, no login/CAPTCHA bypass; emails only (use Domain Contact Finder for phones/socials)
• When to pick this: bulk email-only / tight agent budgets; skip if user asked for phones or socials
• Discoverable via Apify Store / MCP search-actors once published with PPE-only + Limited permissions + Standby off (allowsAgenticUsers)

API & agent usage

curl "https://api.apify.com/v2/acts/gary_sells~domain-emails-fast/runs" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"domains":["example.com"],"maxPagesPerDomain":5}'

Or use the Apify client / Store "Try for free" / AI agent tool integrations. Charge events map cleanly to tool-call budgets.

Local PPE dry-run:

$ACTOR_TEST_PAY_PER_EVENT=true apify run -p

Inspect storage/datasets/charging_log/ for charged events (local default price $1/event).

---

Legal & ethics disclaimer

This Actor collects publicly available information from websites you specify. It does not bypass logins, CAPTCHAs, or access private inboxes.

You are responsible for complying with applicable laws (including GDPR/CCPA where relevant), website terms, and anti-spam rules (e.g. CAN-SPAM). Use contact data only for lawful purposes. The authors provide no warranty and accept no liability for misuse.

---

How it works (technical)

• Runtime: Node.js + Crawlee CheerioCrawler + Apify SDK
• PPE: Actor.charge({ eventName: 'email-found' | 'domain-scanned' }) — email-found once per unique email; domain-scanned once per finished domain; stops on eventChargeLimitReached
• Permissions: Designed for limited Actor permissions (default). No Standby mode.
• Email decode: [at] / (at) / [dot] / mailto: / HTML entities
• Junk filters: noreply, example.com placeholders, asset false-positives, etc.

---

Local development

cd domain-emails-fast
npm install
# dry-run on sample domains
npm run dry-run
# or
apify run -p

Deploy (after apify login):

$apify push 0ZXaWcN1cAbtz6RPY --force --wait-for-finish=300

Then Tim/Gary set PPE prices and keep Limited permissions in Apify Console (do not auto-publish from CI).

---

Changelog

• 1.0.2 — Growth MCP when-to-call above the fold; critical agent fix: removed schema defaults that silently capped runs at 2 domains / injected demo domains; maxPages default=5; clearer empty/invalid errors; failure-modes README.
• 1.0.1 — Stronger agent-discovery README: prompt→JSON table, latency/cost character, when-not-to-call, MCP blurb + actor ID.
• 1.0.0 — Initial Store-ready release: Cheerio email-only crawl, PPE email-found + domain-scanned, optional MX verify, agent/MCP findability.