Meta Tag Extractor
Title, description, canonical and social tags from any page. One at a time or twenty at once.
What the length figures mean
Titles and descriptions are cut in search results by pixel width, not by character count. A title in wide capitals runs out of room sooner than the same number of narrow lowercase letters, so any character limit is an approximation.
| Tag | Working figure | What happens past it |
|---|---|---|
| Title | About 60 characters | Cut with an ellipsis. The important words should come first anyway. |
| Description | About 155 characters | Cut, or replaced entirely with a passage from the page. |
The counts above are marked green, amber or red against those figures. Amber means it will probably survive; red means it will almost certainly be trimmed.
Your description is a suggestion, not an instruction
Search engines rewrite meta descriptions whenever they judge that a passage from the page answers the query better, which for long-tail searches is most of the time. That is not a reason to leave it blank: it is used often enough to matter, particularly for brand and homepage searches. It is a reason not to spend an afternoon on one.
What is read, and what is not
The page is fetched as an anonymous visitor, with no cookies and no JavaScript. That is deliberately close to what a crawler sees on its first pass. If your tags are injected by a script after load, they will not appear here, and that is worth knowing rather than working around, because it means they may not be seen reliably in search either.
Why a private address is refused
A tool that fetches whatever URL a stranger types is, without care, a way to make our server issue requests on their behalf. Someone submits http://127.0.0.1/, or a cloud metadata address, and gets back something that was never meant to leave the private side of a network. The problem has a name, server-side request forgery, and it is one of the more common ways a helpful tool becomes a liability.
Addresses on private, loopback, link-local and reserved ranges are therefore refused before any connection is made. The hostname is resolved first and every address it returns is checked, the connection is pinned to the address that was checked so the name cannot answer differently a moment later, and redirects are followed one hop at a time with the same checks applied to each. Only http and https are accepted; a file: address is not a URL this will read.
Open Graph and Twitter tags
These control how a link looks when it is shared rather than how it appears in search, and they are the ones most often forgotten. A page with no og:image shares as a bare line of text on every platform; a page with no og:title falls back to the page title, which is usually fine but occasionally wrong. Both are shown above where present, and their absence is called out rather than left blank.
How to use it
Enter an address and press Extract. Switch to Bulk to paste up to twenty addresses, one per line, and export the lot as CSV. A bare domain works; the scheme is added if you leave it out.
Questions
How long should a title and description be?
Around 60 characters for a title and 155 for a description are safe working figures. Neither is a hard limit: search engines cut by pixel width rather than by character count, so a title of wide capitals truncates sooner than one of narrow lowercase letters. Treat the numbers as a guide and check the preview.
Why does Google show a different description from mine?
Because the meta description is a suggestion rather than an instruction. Search engines rewrite it whenever they think a passage from the page answers the query better, which is most of the time for long-tail searches. It is still worth writing well, since it is used often enough to matter.
Can this read a page behind a login?
No. The page is fetched as an anonymous visitor with no cookies and no credentials, so anything requiring a session returns the login page instead. That is deliberate: a tool that could reach authenticated pages would be a tool for reading other people's private pages.
Why can it not read a page on my local machine?
Because addresses on private, loopback and reserved ranges are refused before any connection is made. Allowing them would turn this into a way of making our server probe networks on a stranger's behalf, which is the well-known server-side request forgery problem.
Does it run JavaScript on the page?
No. It reads the HTML the server sends, which is also what a crawler sees first. If your tags are inserted by JavaScript after load they will not appear here, and that is worth knowing rather than working around: it means they may not be seen reliably in search either.
Related tools
Browse all Sigma Wire tools - every free tool on the site, grouped by category.