New set of weak CAPTCHA

4/5 (1)

After some time of browsing the web we found a set of new CAPTCHAs which must take a place in a Weakness list. Some of them are small and local, but it also contain a world wide famouse systems. Due to a self-confidence of webmasters we will have a lot of places where OCR may be successfully applied.


4/5 (1)

One more mobile operator – Jeans. Guys who made this CAPCTHA was inspired by Rapidshare CAPCTHA. But they don’t trace it’s changings – such model was hacked very quickly. Due to color model separating to chars is very easy. Then neural network makes the rest.

version 1.2

3/5 (1)

Here is new stable and good version of tEABAG_3D. It’s rewritten in PHP5, and works ONLY WITH PHP5. It’s now completely OOP well-made, must be much easier to implement and use. Image if live, it will change with each refresh. As it was promised, all buyers of previous versions can get it for free. Contact us.

Kyivstar SMS sender may be automated

2/5 (1)

Our Team has completed OCR for Kyivstar SMS sender. It shows, that Kyivstar don’t want to spent even small money to protect customers from available SMS spam attack. They will think of you There is an opinion, that they will not change this system until SMS spam starts.

UMC (ver 0.2)

3/5 (1)

New version of UMC CAPTCHA is much better then previous (see demo). Char overlapping is great. BUT – it’s random there, so, sometimes there is no overlaping at all. And, of course, the char color. Thank you very much for making overlapped chars with different colors. Due to this feature overlapping means nothing, and gives no additional obfuscation. Must say that readability went down – it’s now not too easy to read it correctly.

4/5 (1) is a new giant, guys. But not in CAPCTHAs. Yes, must say, it’s nice, and hard to beat, but color model is not the only way to separate specific chuncks. It’s possible to separate letter chuncks from background chunks basing in it’s parameters, firstly – position and shape.