How it works
The tool finds text that starts with http://, https:// or www. and continues until a space, quote or angle bracket. Then it trims characters that usually belong to the surrounding sentence rather than the link: a final full stop, comma, colon, semicolon, question mark or closing quote.
Closing brackets are handled carefully. A “)” at the end is removed only if the URL does not contain a matching “(”, so Wikipedia-style links such as …/Foo_(bar) survive intact.
Rules this tool follows
- Duplicates are removed by default, comparing the full URL exactly.
- Bare domains without http or www (“example.com”) are not extracted, because they are too easily confused with file names and abbreviations.
- Turn off “Include links starting with www.” to keep only links with an explicit protocol.
Example
Input
Docs are at https://example.com/docs?page=2, and the wiki (https://en.wikipedia.org/wiki/Plain_text) explains more. Mirror: www.example.org.
Output
https://example.com/docs?page=2 https://en.wikipedia.org/wiki/Plain_text www.example.org
Common uses
- Collecting links from emails, chat exports and documents.
- Checking which external sites a piece of content references.
- Preparing a list of URLs for review or link checking.
Limitations and notes
- A URL that genuinely ends in a full stop or comma will lose it.
- Links inside HTML attributes are found only if they start with http or www; relative links such as /page are not. Convert HTML to text first with HTML to Text and “Show link URLs”.
- Percent-encoded links are extracted as written; decode them with the URL Decoder.