Tool guideEmail Tools

Email Address Extraction – Guide to Finding Addresses in Text and Websites

What email address extraction is and how it differs from verification, where addresses hide in HTML, how obfuscation and IDN domains are handled, what crawl depth and robots.txt mean, and how to use the extractor responsibly.

By NetTools.tech · · 5 min read

Email addresses end up in many places: contact pages, footers, PDFs exported to HTML, old log files, spreadsheets and email headers. Finding them by hand is slow, and a simple search misses addresses that are encoded, obfuscated or hidden in metadata.

The Email Address Extraction tool finds them for you. Text and files are processed in your browser, and public websites are scanned by a restricted crawler that respects robots.txt. The results are de-duplicated and grouped by domain and source.

What is email address extraction?

In simple terms

Extraction means scanning content for text that has the shape of an email address, such as [email protected], and listing every one it finds, once, with where it came from.

Technical background

An address has a local part, an @ and a domain. The extractor accepts the common dot-atom local part (letters, digits, . _ % + -) and domains with Unicode labels (IDN) or their punycode form (xn--…), with any top-level domain — not just .com. It rejects strings that only look like addresses, such as retina image names (logo + @2x + .png) or [email protected] version strings. Each result is checked against the length and label rules of RFC 5321 and marked as valid syntax, invalid syntax or unusual format.

Why use an email extractor?

  • Find every address your own website publishes, including ones you forgot about.
  • Pull the addresses out of exported CSV files, logs or configuration before a migration.
  • Read the senders and recipients of an .eml file without opening it in a mail client.
  • Check whether obfuscation on your contact page really hides an address from simple scrapers.

Who should use it?

Website owners and administrators auditing their own sites, developers cleaning up data, system administrators reviewing logs, and security researchers checking what a public site exposes about its staff.

How the email extractor works

Text & Files reads pasted text and files (TXT, HTML, CSV, EML, LOG, JSON, XML, Markdown) locally. The input type is detected automatically or chosen manually.

Website URL and Multiple URLs send the URLs to the NetTools crawler. For each page, the crawler:

  1. Checks robots.txt.
  2. Fetches the page from a public address only, on port 80 or 443.
  3. Follows redirects one hop at a time, re-checking each destination.
  4. Extracts addresses.
  5. Queues the same-site links it found, breadth-first, up to the chosen depth.

It waits between requests to the same host, honours Crawl-delay, and never runs JavaScript.

In HTML, addresses are found in these places:

  • Visible text.
  • mailto: links. A mailto: link is a hyperlink that opens a mail program, for example <a href="mailto:[email protected]">.
  • Attributes and comments, labelled as HTML source.
  • <meta> tags and JSON-LD structured data.
  • Text written with HTML entities, such as info&#64;example&#46;com.
  • If you enable it, common obfuscations such as name [at] example [dot] com.

How to use the NetTools.tech Email Extractor

  1. Open Email Address Extraction and choose a tab.
  2. For text, paste content or drop files and press Extract email addresses.
  3. For a website, enter the URL, pick a crawl depth and page limit, and press Scan website. Watch the progress, and cancel at any time.
  4. Filter the results by domain, source, depth or type, and switch between unique addresses and all occurrences.
  5. Copy addresses, or export them as PDF, CSV or JSON with the toolbar.

Understanding the results

Unique addresses are compared case-insensitively, with the domain converted to punycode. Local parts are never merged by provider-specific rules: john.smith@ and johnsmith@ stay different. Occurrences count every place an address appeared. The domain analysis groups addresses by domain. The source pages table shows each page's status, title, links found and followed, and why links were skipped.

Example: auditing a company contact page

A scan of a company site at depth 1 finds 14 occurrences of 5 unique addresses:

  • info@ appears on every page through a footer mailto: link.
  • press@ is written as press&#64;company.example on the press page, so it shows as HTML-encoded.
  • jobs [at] company [dot] example is found only with obfuscation detection on.

The source-page table shows that /internal/ was blocked by robots.txt and that two links led to an external booking site, so they were not followed.

Common use cases

  • Inventory of published addresses before a domain or mail migration.
  • Checking that a privacy request removed an address from every page.
  • Extracting senders from saved emails or headers.
  • Cleaning address columns in exported spreadsheets.

Troubleshooting

The website scan finds nothing

The page may build its content with JavaScript, show addresses as images, or require a login. Check the source-page table for robots.txt blocks, redirects to other sites, or HTTP errors.

Pages are blocked by robots.txt

The crawler obeys the site's robots.txt for NetToolsBot (or *). If robots.txt cannot be fetched because of a server error, the crawler treats the whole site as disallowed.

The scan stopped early

A limit was reached: page count, scan duration, download size or the number of addresses. Narrow the scan or lower the depth.

Some sites cannot be scanned

Bot protection, rate limits or firewalls may refuse the crawler. The tool reports this and does not try to get around it.

Common mistakes

  • Treating valid syntax as proof that a mailbox exists.
  • Assuming every provider ignores dots and plus tags in addresses.
  • Scanning third-party websites and then mailing the addresses without consent.
  • Expecting addresses that JavaScript inserts after the page loads to be found.

Limitations

The crawler does not execute JavaScript, read PDFs or images, or log in. Robots.txt support follows RFC 9309 (groups, Allow/Disallow, wildcards, Crawl-delay). Relaxed subdomain grouping uses an approximation of the registrable domain. Server limits, such as depth, pages, time and size, are set by the administrator.

Security & privacy

Pasted text and files never leave your browser. Website scans fetch only public addresses on standard ports, and each redirect and DNS answer is checked against private and internal ranges. Results are kept in server memory only for the scan and deleted after a few minutes. URLs, page contents and addresses are not logged. Administrators see only anonymous counts.

Summary

Email extraction finds the addresses in text, files and public web pages. It catches mailto links, encoded addresses and obfuscated ones. NetTools.tech does this locally for your own content and with a polite, robots.txt-respecting crawler for websites. It checks syntax only and never verifies or contacts a mailbox.

Frequently asked questions

Does the extractor check whether an address exists?

No. It checks syntax only and never contacts mail servers. A valid-looking address can still be unused, misspelled or deliberately fake.

Is my pasted text or uploaded file sent to NetTools?

No. Text & Files mode runs entirely in your browser. Only website scans use the NetTools crawler, which keeps results in memory for a few minutes and never stores them.

Why does the crawler miss some addresses on a site?

Addresses created by JavaScript, shown as images, placed behind a login or on pages that robots.txt disallows are not found. The crawl also stops at the chosen depth and page limit.

What does crawl depth 1 mean?

Depth 0 scans only the URL you enter. Depth 1 also scans the pages that this URL links to on the same site. Depth 2 adds one more level of links.

Are [email protected] and [email protected] treated as the same address?

No. Some providers ignore dots or plus tags, but that is not universal, so the extractor only compares addresses case-insensitively and never merges local parts.

May I collect addresses for a mailing list?

Sending unsolicited mail to collected addresses breaks anti-spam and privacy laws such as GDPR and CAN-SPAM. Use the tool to audit your own sites and documents.