HTML Link Extractor

Pull every link out of a page, split internal from external, and spot empty or generic anchor text.

How to use

What it checks and how it works

The tool reads every a element with an href in the pasted HTML. A link is internal when its host matches the site address you give, or the base or canonical address in the HTML if you give none, and external when it does not. Links with mailto, tel or javascript schemes and in-page anchors are counted apart. Subdomains can be counted as internal using the registrable domain from the tldts package. It lists anchor text and rel values (nofollow, sponsored, ugc) and flags empty anchors, generic anchors such as click here, repeated destinations and a elements with no href. The rules follow Google Search Central: a link is crawlable when it is an a element with an href, anchor text should be descriptive, and nofollow, sponsored and ugc are the documented rel values for qualifying outbound links. The English generic anchors are the examples Google gives; the Arabic list is Zaher's equivalent. Limits: it reads only the HTML you paste, so links added by JavaScript appear only if you paste the rendered HTML. It requests no address, so it cannot tell whether a link works, redirects or is blocked.

Does it check whether links are broken?

No. The tool never fetches an address, which keeps your page private, so use the Redirect Chain Checker to follow a redirect.

What counts as an internal link?

A link whose host matches the site address you enter, or the page's base or canonical address if you leave it empty. Tick the subdomain option to count subdomains of the same registrable domain.

Which rel values does Google document?

Google documents nofollow, sponsored and ugc for qualifying outbound links. The tool counts each of the three.

Sources

Related tools

Privacy

Runs in your browser; nothing is uploaded. The interactive tool needs JavaScript.