AnyTool
Your files never leave your device. All processing happens locally in your browser.

How do I extract all URLs from text?

Paste your text, HTML or markdown into AnyTool URL Extractor and every link is pulled out instantly. It catches http(s) and ftp links, www. hosts (auto-prefixed with https://), optional bare domains, mailto:/tel:, markdown [text](url) links and HTML href/src attributes. Each URL is parsed into its protocol, host, port, path, query parameters and hash, and you can normalise, strip tracking parameters, dedupe, sort and filter by host. Everything runs in your browser, so nothing is uploaded.

  • Live extraction as you type or paste, from plain text, HTML or markdown
  • Finds http(s)/ftp links, www. hosts, bare domains, mailto:/tel:, markdown and href/src
  • Parses each URL into protocol, host, port, path, query params and hash
  • Normalise hosts, strip utm_*/gclid/fbclid tracking, dedupe, sort and validate
  • Per-host breakdown plus copy or download as newline, comma, markdown or CSV — 100% client-side

What is

URL Extractor

A URL extractor is a tool that scans a block of text, HTML or markdown and pulls out every string that looks like a link, then parses each one into its components with the browser URL engine so the results can be cleaned, deduped, grouped by host and reused as a list.

Text Tools

Related terms

Link FinderURL ParserTracking Parameter RemovalEmail Extractor

Frequently Asked Questions

No. All extraction and parsing run in your browser, so your text never leaves your device and the tool works offline.

Every step — finding URLs, parsing them with the native URL constructor, normalising, stripping tracking parameters, deduping and the host breakdown — happens locally in your browser using JavaScript. Nothing is sent to a server, there is no tracking, and the page works offline after the first load, so you can safely paste confidential pages, exports or logs.

Yes. It reads href, src, srcset and similar attributes from HTML and pulls the URL out of markdown [text](url) and image links.

Beyond plain text, the extractor surfaces links hidden in markup: HTML attributes such as href, src, srcset, action and poster, and markdown link and image syntax like [docs](https://example.com) or ![logo](https://example.com/a.png). Those URLs are harvested first, then a free-text scan catches the rest — schemed links, www. hosts and, if enabled, bare domains.

Yes. Turn on Strip tracking to remove utm_* parameters plus gclid, fbclid, msclkid and other common click identifiers.

The Strip tracking option rewrites each URL to drop the utm_ family (utm_source, utm_medium, utm_campaign and so on) along with ad and analytics identifiers such as gclid, dclid, fbclid, msclkid and yclid, while keeping the genuine query parameters. Combined with the Normalise option, which lowercases the host and removes a trailing slash, it produces clean, canonical links.

Yes. Filter to keep or exclude listed hosts, then copy or download the result as a newline, comma, markdown or parsed-column CSV.

Enter one or more hosts to keep only those links or to exclude them — subdomains match too, so acme.io also matches blog.acme.io. The cleaned list can be sorted and exported as a newline list, comma-/space-separated text, a markdown bullet list, or a CSV whose columns are the parsed url, protocol, host, port, path, query and hash, all generated in your browser.

Detailed Explanation

Methodology

How the URL Extractor Works

AnyTool URL Extractor processes text entirely in the browser. It first harvests links embedded in markup — HTML href, src, srcset, action and poster attributes, and markdown link and image syntax such as [docs](https://example.com) — surfacing them as plain tokens. A free-text scan then catches schemed links (http, https, ftp, ftps), www. hosts that are auto-prefixed with https://, and, optionally, bare domains and mailto:/tel: links. Trailing sentence punctuation and unmatched closing brackets are trimmed, and each surviving URL is parsed with the native URL constructor.

  • Harvests href/src/srcset attributes and markdown link and image URLs
  • Scans for http(s)/ftp links, www. hosts and optional bare domains
  • Optional mailto: and tel: link extraction
  • Trailing punctuation and unbalanced brackets are trimmed off matches
  • All processing is client-side JavaScript — no server round-trip
How It Works

Parsing, Normalisation and Tracking Removal

Each URL is broken into its protocol, host, port, path, query parameters and hash using the browser URL engine, with query parameters listed key by key. A normalise option lowercases the host and removes a trailing slash, while a strip-tracking option removes the utm_ family along with gclid, dclid, fbclid, msclkid, yclid and other ad and analytics click identifiers, leaving genuine parameters intact. The list can then be deduped, validated, sorted A→Z, Z→A or grouped by host, and filtered to keep or exclude specific hosts with subdomain matching.

  • Native URL parsing into protocol, host, port, path, query and hash
  • Normalise lowercases the host and strips a trailing slash
  • Strip tracking removes utm_*, gclid, fbclid, msclkid and similar params
  • Dedupe, validate, and sort A→Z, Z→A or by host
  • Per-host breakdown with counts and headline totals
Privacy & Security

Privacy and Offline Use

Because extraction and parsing run in the browser, the source text never reaches a server, there is no account or tracking, and the page works offline after first load. Large inputs are debounced so the editor stays responsive, and results can be exported as a newline list, comma- or space-separated text, a markdown bullet list, or a CSV with parsed columns — making it safe to extract links from confidential pages, CRM exports or server logs.

URL extraction: in-browser (AnyTool) vs typical online extractors
CapabilityAnyToolTypical online extractors
Text handlingProcessed in your browserOften posted to a server
PrivacyText never leaves deviceMay log or store submitted text
Source coverageText, HTML attrs & markdownOften plain text only
URL parsingProtocol/host/path/query/hashRarely built in
Tracking removalutm_*, gclid, fbclid & moreRarely offered
Output formatsNewline, comma, markdown, CSVOften newline only
Works offlineYes (PWA)No
Cost / signupFree, no signupOften ad-heavy or gated

Extraction is pattern-based and parses links locally; AnyTool never fetches a URL and uploads nothing.