Also: the Internet Archive's crawler (Heritrix) implements link extractors for HTML, XML, CSS, JS, PDF, Word document, and even Shockwave Flash.

Heritrix extractor Javadocs: https://www.javadoc.io/doc/org.archive.heritrix/heritrix-mod...

Source of the JavaScript extractor: https://github.com/internetarchive/heritrix3/blob/master/mod...

A script is included for running the extractors from the command line: https://github.com/internetarchive/heritrix3/blob/master/dis...

Says it's inspired by "LinkFinder", which was useful in dredging up what the original purpose of the tool was: https://gerbenjavado.com/discovering-hidden-content-using-li...

If it isn't a bit late, you might want to reconsider the name as it clashes with an existing web project: https://en.wikipedia.org/wiki/Links_(web_browser)

Not a commonly used one in my experience, but still actively maintained.

Which ironically also clashes with https://en.wikipedia.org/wiki/Lynx_(web_browser

I seem to recall that one was actually deliberate, though I can't find a reference that says that ATM.

Its spelled differently and in a different market/domain, I think it should be fine

What does "invisible links" mean? Obfuscated JS? An example would be helpful in the readme.

It says it's inspired by LinkFinder which describes itself as:

"LinkFinder is a python script written to discover endpoints and their parameters in JavaScript files. This way penetration testers and bug hunters are able to gather new, hidden endpoints on the websites they are testing."

Thanks, It said inspired by, I didn't know if it did more than that.

It's probably confusing "links" and "URLs."

Thanks a lot for the suggestions, the ratings make me very happy! I approach all comments positively, I will consider them all. <3

Do we have a FF or Chrome extension that can reveal invisible links within HTML pages?

I would absolutely use this if it existed.

Cool! I was looking for something exactly like this before when reverse-engineering some website's API.

A few suggestions:

The current algorithm seems to have way too many false positives, it got things like "YYYY/M/D", "Asia/Hong_Kong|Hongkong", "video/x-matroska" etc. Not a deal breaker.

The main pain point when using it is the way it presents the result.

It firstly has quotes around each result, making copy them quite annoying (double click would select these quotes too).

Secondly, I wish it would automatically attempt to generate a link using relative path so I can directly click and test. It can be a separate column with <a>.

Lynx/Links is a browser. Name's taken...