Email Extractor
Paste text or HTML and get every address — deduplicated, grouped and ready to export.
Works only on the text you give it — nothing is uploaded and no website is visited.
This looks like a web address. The extractor does not visit websites: open the page, select all its text (or view its source with Ctrl+U), copy it and paste it here.
What to find
Separate domains with commas or spaces. Subdomains are included: example.com also matches mail.example.com.
Addresses
By domain
| Domain | Addresses | Mentions |
|---|
Not included
| Text | Why it was left out | Times |
|---|
About the Email Extractor
Paste any text — an email thread, a web page’s HTML source, a CSV export, a signature block — or open a text, CSV, HTML or .eml file, and every email address in it is listed once, however many times it appears. Addresses in mailto: links, addresses hidden with HTML codes such as @, and, if you want, forms like name [at] example [dot] com are found too. Each address is checked against the email standards, display names such as Priya Sharma <[email protected]> are kept for the CSV, and the results are grouped by domain.
The extractor works only on the text you give it: it never visits web pages and nothing you paste is uploaded, so customer lists and private mail stay on your device.
How to use it
- Paste your text or HTML, or use Open file for a
.txt,.csv,.htmlor.emlfile (drag and drop works too). - Choose what to look for: hidden forms such as name [at] domain [dot] com, rare symbols before the @, and whether only real top-level domains count.
- Optionally keep or leave out domains (for example your own company’s), and skip no-reply addresses.
- Pick the order and the format — one per line, comma-separated, semicolon-separated (for Outlook) or CSV with names — then Copy or Download.
Examples
To: "Rao, Arjun" <[email protected]>, Priya Sharma <[email protected]> Cc: [email protected]
[email protected] is the same address, so it is counted as a second mention. The CSV keeps the names “Rao, Arjun” and “Priya Sharma”.
<a href="mailto:hello%40example.net">Mail us</a> · sales [at] example [dot] org · jo@example.com
[email protected] · [email protected] · [email protected]:org/repo.git
(no addresses)
[email protected] is listed under “Not included” because .png is a file type, not a top-level domain.
Common uses
- Collecting the participants of a long email thread or a meeting invite.
- Pulling contact addresses out of an exported CRM, spreadsheet or website’s HTML source.
- Cleaning a messy contact list: one address per line, duplicates removed, sorted by domain.
- Making a semicolon-separated recipient list to paste into Outlook’s To, Cc or Bcc field.
What counts as an email address
An address is the part before the @ (the local part) and a domain. The extractor follows the email standards with a few practical limits:
- Before the @: letters from any script (RFC 6531 allows international addresses), digits and
. _ - + '. The other symbols that RFC 5322 allows (! # $ % & * / = ? ^ { | } ~and the backtick) are legal but almost never used, and they turn text such as?email=…in links into fake addresses, so they are off unless you turn them on. Dots may not start, end or appear twice in a row, and the part is limited to 64 bytes (RFC 5321). - The domain: at least two parts separated by dots, each 1–63 characters of letters, digits and hyphens, not starting or ending with a hyphen. International domains such as bücher.de are checked with the same rules browsers use for web addresses (WHATWG URL Standard), and bücher.de and its Punycode form xn--bcher-kva.de count as the same domain.
- The ending: with Only real top-level domains on, the last part must be a top-level domain that exists, such as .com, .in, .uk or .рф. This rejects file names like [email protected] and private names like .local; turn it off for internal addresses.
- Length: at most 254 characters in total.
Capital letters do not matter when finding duplicates. The domain is shown in lower case; the part before the @ is kept as first written.
These checks are stricter than the HTML standard’s “valid email address” used by <input type="email">, which deliberately allows two dots in a row and an address without a dot in the domain (such as user@localhost), and they also accept international letters, which that definition does not.
Hidden and encoded addresses
- mailto: links are read even when the address is percent-encoded (
john%40example.com), written as Name <address> or lists several recipients, includingcc=andbcc=. Outlook’s quoted reply headers, such as From: Doe, Jane [mailto:[email protected]], keep the name for the CSV. - HTML character codes such as
@,@and@are decoded before searching. - Email source (.eml) marked as quoted-printable is decoded, so addresses broken across lines with
=are joined again. - Hidden forms: name [at] example [dot] com, with round, square, curly or angle brackets. Plain words (name at example dot com) can be turned on too, but ordinary sentences sometimes match, so check the result.
Not extracted on purpose: version numbers ([email protected]), SSH remotes ([email protected]:org/repo.git), user names and passwords in links (ftp://user@host, https://user:token@host) and handles without a domain (@name).
Limitations
- Only the format is checked. Whether a mailbox really exists can only be found out by a mail server, which this tool does not contact.
- Rare forms are not extracted: quoted local parts ("john smith"@example.com) and IP-address domains (user@[192.0.2.1]).
- Addresses inside images, or split across lines in ordinary text, cannot be found. For a PDF, get its text first with PDF to Text.
- In Chinese, Japanese, Thai and other text written without spaces, an ordinary address that touches the words around it is cut where the script changes (邮箱[email protected] gives [email protected]). An address whose part before the @ is itself written in such a script cannot be told apart from the words just before it.
- Files up to 50 MB can be opened; extremely large pastes may take a few seconds.
Privacy
Everything happens in your browser. What you enter or open here is not uploaded or stored by MySmartCoPilot.
Frequently asked questions
How do I extract email addresses from a website?
This tool does not visit websites. Open the page in your browser, select all its text (Ctrl+A, or ⌘A on a Mac) and copy it, or view its source (Ctrl+U) and copy that, then paste it here. Pasting the HTML source also finds addresses in mailto: links.
How do I get email addresses out of an Excel or CSV file?
Save or export the sheet as CSV and open it with Open file, or copy the cells and paste them. Every address in any column is found.
Does it remove duplicate addresses?
Yes. Each address is listed once, ignoring capital letters, and Times found in the CSV says how often it appeared. Sort by domain to see addresses from the same company together.
How do I paste the list into Outlook or Gmail?
Choose Semicolon-separated for Outlook, or Comma-separated for Gmail and most other mail apps, copy, and paste into the To, Cc or Bcc field.
Why was an address not included?
Open Not included under the results: it lists every rejected candidate with the reason, such as two dots in a row or a file type instead of a domain. Addresses outside your domain filters, or no-reply addresses when you skip them, are counted there too.
Can I send marketing email to every address I extract?
Not necessarily. In many countries, data-protection and anti-spam laws only allow marketing email to people who have agreed to receive it. Check the rules that apply to you before using a list.
Is my text uploaded?
No. The addresses are found in your browser; nothing you paste or open is sent to a server or stored.