Email Syntax Validator

Email Syntax Validator splits each entry at its last @ and tests the two halves separately. The local part must be no longer than 64 characters, must not start or end with a dot or carry two in a row, and must be built from the RFC 5322 atext set — letters, digits and the punctuation the grammar actually permits — so a parenthesis comes back named as an invalid character, while a space anywhere outside a quoted local part fails earlier still, as whitespace, before the two halves are examined at all. The domain must be at most 253 characters across labels of at most 63, with no leading, trailing or doubled dots, no label beginning or ending with a hyphen, and a top-level label that is either two or more letters or a punycode xn-- label. An all-digit top-level domain is refused, and the whole address is capped at 254 characters.

It also handles the shapes that arrive in real lists rather than in specifications. A display-name entry such as Bob Smith <[email protected]> is unwrapped to the address inside and noted as such, a leading mailto: is stripped, a quoted local part is accepted with a note that it is valid but widely unsupported, and a bracketed IP literal is accepted with the same caveat. An internationalised domain written in its Unicode form passes and is flagged; a non-ASCII local part passes with a note that delivery needs SMTPUTF8 support at both ends. Alongside those, the Detail column marks role addresses like info@ and noreply@, plus-addressing, and local parts or domains that look like placeholders.

The Did you mean column compares each domain against a fixed list of forty-three consumer providers using edit distance, and against a short table of top-level misspellings — con, cmo, ocm and vom for com, ogr for org, nte for net. This is deliberately kept separate from the verdict. [email protected] is a perfectly legal address by syntax, so it is marked valid with a warning and a suggested correction rather than rejected, and the tool never invents a fix for a domain it already recognises. Those two halves have different reach, which is worth knowing before you trust a blank column. The misspelling table rewrites the last label of any two-label domain, so [email protected] is corrected to .com even though that company is on no list; the provider list only catches a domain within an edit or two of one of the forty-three, so a garbled company name like [email protected] passes with nothing suggested. Add a subdomain and even the table goes quiet — [email protected] reads as plainly valid.

The limit that matters is what syntax can tell you, which is less than most people want. There is no DNS query, no MX record check and no SMTP probe here, so a flawlessly formed address at a domain that has never existed passes without comment — use MX Lookup to find out whether a domain accepts mail at all. Two rules will also surprise you. user@localhost and anything else with a single-label domain is refused, because the check requires a top-level domain, even though a local mail system would deliver it. And the input is split on commas and semicolons as well as line breaks, so a display name containing a comma is torn into two rows — put one address per line whenever names are involved. Deduplication lowercases the entire address, including the local part that RFC 5321 formally treats as case-sensitive, so Bob@ and bob@ collapse into one row. All of this runs inside your browser tab; the list is never uploaded, which matters because a column of addresses is personal data about other people, not just a file of your own.

Runs on your device. Your file is processed directly in this browser.