Update: Prize information
Each year the web security community produces a stunning amount of new hacking techniques documented in white papers, blog posts, magazine articles, mailing list emails, etc. Within the thousands of pages are the latest ways to attack websites, web browsers, web proxies, and so on. We are NOT talking about individual vulnerabilities with CVE numbers, nor any particular system compromise, but the actual new methods of Web-based attack. To keep track of all these discoveries and encourage information sharing, the Top Web Hacking Techniques acts as both a centralized knowledge base and a way to recognize researchers who contribute excellent work.
The selection process for 2010 will be a little different. Last year in 2009, where over 80 new attack techniques were recorded, the winners were selected solely by a panel of panel of distinguished security experts. This year we'd like you, the Web security community, to have the opportunity to vote for your favorite research. From the voting results the most popular 15 entries will be those judged by our panel of experts on the basis of novelty, impact, and overall pervasiveness to decide the Top Ten Web Hacking Techniques of 2010. Researchers topping the 2010 list may expect to receive praise amongst their peers as have those in past years (2006, 2007, 2008, and 2009). Right now I’m working on a really cool set of prizes for #1.
Then at IT-Defense 2011 (Feb.) it will be my great honor to introduce each of the top ten during my “Top Ten Web Hacking Techniques of the Year (2011)” presentations. Each technique will be described in technical detail for how they function, what they can do, to whom, and how best to defend against them. The audience will get an opportunity to better understand the newest Web-based attacks believed most likely to be used against us in the future.
To make all this happen we are going to need a lot of help from the community. At the bottom of this post will be the living master list of everything recorded. If anything is missing please comment containing the link to the research. Or maybe you think something should not be on the list. That's cool, but please explain why. While clearly not every technique is as powerful as another, please make every effort to include them anyway. Nothing should be considered too insignificant. Sometimes several issues can be combined for amazingly effective techniques.
Thank you!
Prizes:
1) OWASP Conference Pass
2) Autographed copies by the authors of "Hacking: The Next Generation", "Hacking Exposed Web Applications 3rd Ed", "24 Deadly Sins of Software Security", "XSS Attacks: Cross Site Scripting Exploits and Defense", "Foundations of Security", "Hacking Web Services", "Web 2.0 Security", "Web Application Obfuscation", "Seven Deadliest Web Application Attacks", "ModSecurity Handbook", "Apache Security", "The Web Application Hacker's Handbook", "SQL Injection Attacks and Defenses", "Detecting Malice", and "Web Security Testing Cookbook."
3) BlackHat USA 2011 Conference Pass
The Complete List of Attack Techniques
- Evercookie
- Hacking Auto-Complete (Safari v1, Safari v2 TabHack, Firefox, Internet Explorer)
- Cookie Eviction
- Converting unimplementable Cookie-based XSS to a persistent attack
- phpwn: Attack on PHP sessions and random numbers
- NAT Pinning: Penetrating routers and firewalls from a web page (forcing router to port forward)
- Mapping a web browser to GPS coordinates via router XSS + Google Location Services without prompting the user
- XSHM Mark 2
- MitM DNS Rebinding SSL/TLS Wildcards and XSS
- Using Cookies For Selective DoS and State Detection
- Quick Proxy Detection
- Flash Camera and Mic Remember Function and XSS
- Improving HTTPS Side Channel Attacks
- Side Channel Attacks in SSL
- Turning XSS into Clickjacking
- Bypassing CSRF protections with ClickJacking and HTTP Parameter Pollution
- CSS History Hack In Firefox Without JavaScript for Intranet Portscanning
- Popup & Focus URL Hijacking
- Hacking Facebook with HTML5
- Stealing entire Auto-Complete data in Google Chrome
- Chrome and Safari users open to stealth HTML5 AppCache attack
- DNS Rebinding on Java Applets
- Strokejacking
- The curse of inverse strokejacking
- Re-visiting JAVA De-serialization: It can't get any simpler than this !!
- Fooling B64_Encode(Payload) on WAFs and filters
- MySQL Stacked Queries with SQL Injection...sort of
- A Twitter DomXss, a wrong fix and something more
- Get Internal Network Information with Java Applets
- Java DSN Rebinding + Java Same IP Policy = The Internet Mayhem
- Java Applet Same IP Host Access
- ASP.NET 'Padding Oracle' Crypto Attack
- Posting raw XML cross-domain
- Generic cross-browser cross-domain theft
- One vector to rule them all
- HTTP POST DoS
- Penetrating Intranets through Adobe Flex Applications
- No Alnum JavaScript (cheat sheet, jjencode demo)
- Attacking HTTPS with Cache Injection
- Tapjacking: owning smartphone browsers
- Breaking into a WPA network with a webpage
- XSS-Track: How to quietly track a whole website through single XSS
- Next Generation Clickjacking
- XSSing client-side dynamic HTML includes by hiding HTML inside images and more
- Stroke triggered XSS and StrokeJacking
- Internal Port Scanning via Crystal Reports
- Lost in Translation (ASP’s HomoXSSuality)
- Cross Site URL Hijacking by using Error Object in Mozilla Firefox
- JavaSnoop
- IIS5.1 Directory Authentication Bypass by using ":$I30:$Index_Allocation"
- Universal XSS in IE8
- padding oracle web attack (poet, Padbuster, demo)
- IIS6/ASP & file upload for fun and profit
- Google Chrome HTTP AUTH Dialog Spoofing through Realm Manipulation
- NoScript Bypass - "Reflective XSS" through Union SQL Poisoning Trick
- Persistent Cross Interface Attacks
- Port Scanning with HTML5 and JS-Recon
- Performing DDoS attacks with HTML5 Cross Origin Requests & WebWorkers
- Cracking hashes in the JavaScript cloud with Ravan
- Will it Blend?
- Stored XSS Vulnerability @ Amazon
- Poisoning proxy caches using Java/Flash/Web Sockets
- How to Conceal XSS Injection in HTML5
- Expanding the Attack Surface
- Chronofeit Phishing
- Non-Obvious (Crypto) Bugs by Example
- SQLi filter evasion cheat sheet (MySQL)
- Tabnabbing: A New Type of Phishing Attack
- UI Redressing: Attacks and Countermeasures Revisited
55 comments:
I like GDS's adobe flex apps hacks:
http://www.gdssecurity.com/l/b/2010/03/17/penetrating-intranets-through-adobe-flex-applications/
really good one, thanks romain. added #37
How about Nitesh's recent posts?
https://blogs.sans.org/appsecstreetfighter/2010/11/29/ui-spoofing-safari-iphone/
https://blogs.sans.org/appsecstreetfighter/2010/11/08/insecure-handling-url-schemes-apples-ios/
http://blog.kotowicz.net/2010/11/xss-track-how-to-quietly-track-whole.html xss-track, my humble addition to the list.
Next Generation Clickjacking by Paul Stone is an excellent one:
http://contextis.co.uk/resources/white-papers/clickjacking/
Perhaps you had considered these already but just in case:
1) http://blog.andlabs.org/2010/08/xssing-client-side-dynamic-html.html
2) http://blog.andlabs.org/2010/04/stroke-triggered-xss-and-strokejacking_06.html
@Krzysztof thanks for the submission. added #42.
@lava solid, thanks. added #43 - #45.
I liked the recent exploit for CakePHP:
http://seclists.org/fulldisclosure/2010/Nov/148
It's one of the first public examples of a new attack category against PHP.
@reed: do you have a link to the core issue that explains the "new attack category against PHP"?
When will the voting start/submissions close? I have an as-yet unfinished attack that could still use some polish before I publish it...
@albino: Dec 31. :) Voting will commence shortly thereafter. Tight deadlines for the Feb presentations.
Simple, but extremely effective. Not particularly novel, but a
powerful combination of effective conventional malware and modern
delivery mechanics via persistent script injection -- the return of tweaked malware (i.e. ASPR0x) to the world demands attention.
Additionally, I particularly appreciated the fact that I saw strains of ASPR0x [reincarnated] in which part of the payload was double-encoded to evade detection.
Please check if it is legit.:
ASP’s HomoXSSuality:
http://hackademix.net/2010/08/17/lost-in-translation-asps-homoxssuality/
and
http://soroush.secproject.com/blog/2010/08/noscript-new-bypass-method-by-unicode-in-asp/
@Jason: As cool and powerful as ASPR0x might be, Im am looking for some level of novelty, oh and relative primarily to the Web in some way. I don't think this makes the cut.
@Soroush: HomoXSSuality, that's right, excellent work. #47
Please check if this one is legit. as well:
Cross Site Request Forgery (CSRF) PoC Template (in Javascript)
https://code.google.com/p/csrf-poc-template-by-js/
Although you've said to not post CVE or website issues, there are something that never will be fixed or suggest some methods. I can see some of them in the selected techniques. Please check this out as well:
IIS5.1 Directory Authentication Bypass by using ":$I30:$Index_Allocation"
http://irsdl1.wordpress.com/2010/07/01/iis5-1-directory-authentication-bypass-by-using-i30index_allocation/
@Soroush: tools are cool, but csrf-poc-template-by-js doesn't appear to have a new technique involved. Secondly, would you consider the "IIS5.1 Directory Authentication Bypass" as containing a new filter-evasion technique in there? Something that might apply elsewhere?
You are right about that tool which only makes csrf easier.
I only have 1 thing to say about using Folder:$I30:$Index_Allocation:
I couldn't find any other source that use the same technique to open a directory in win.NTFS. As all of the windows directories are accessible by this technique, it can even bypass some other folder's protection for example in file uploading and so on. I also had written a trick by using similar ADS in the last bullet of section 4 in:
http://soroush.secproject.com/downloadable/Improve%20File%20Uploaders%E2%80%99%20Protections.pdf
Now, it's still up to you. I just wanted to keep this in the list if it is really related.
@Soroush: convincing enough. added #50
I found another method for using :$I30:$Index_Allocation that I cannot release it now ;) But, yes, it can still be dangerous!!!
Padding oracles!! my top 1, actual new method, powerful,high impact, most web dev platforms affected, remote IIS in 2010, JSF, RoR and lots of apps. Thumbs up to all the remote server attcks! too much client side stuff in the list is the same than years ago with a little change.
@ydinoul: I second that.
It's not just ASP.NET. Jeremiah, you should list the paper.
@ydinoul: would much appreciate a link to the best source of information on padding oracles. would be happy to add it.
oki JG! some padding oracle web attack links:
the POET stuff:
http://netifera.com/research/
paper:
http://usenix.org/events/woot10/tech/full_papers/Rizzo.pdf
Padbuster
http://www.gdssecurity.com/l/b/2010/09/14/automated-padding-oracle-attacks-with-padbuster/
demo against dotnetnuke:
http://www.youtube.com/watch?v=yghiC_U2RaM
And like 4 or 5 2010 CVE's but I don't post them! :)
Universal XSS in IE8... http://p42.us/ie8xss/
@ydinoul: added #51. thank you
@tarball: great one, #52
Please look at this:
http://blog.48bits.com/2010/09/28/iis6-asp-file-upload-for-fun-and-profit/
I wish it was in English; However, this guy has written about a new technique in IIS6 which is another Microsoft FeaTure-Bug (FeaFlase!!!) :P
/Folder.asp/test.txt
test.txt will be run as a ASP file!
HTTP://fuzzdb.googlecode.com
Open database of attack patterns and predictable resources
Persistent Cross Interface Attacks -
http://secniche.blogspot.com/2010/11/malware-paradox-cia-aavar-2010.html
Bypassing NoScript 2.0.6 or < - through Union SQL Poisoning using SQLXSSI
http://www.youtube.com/TheKn0ck#p/a/u/1/r-kgKNspqjQ
Web Widgets - Interface Flaws http://www.slideshare.net/adityaks/owasp-app-sec-us-2010
Google Chrome Authentication Dialog Spoofing through Realm Manipulation
http://zeroknock.blogspot.com/2010/08/google-chrome-http-auth-dialog-through.html
Pwning Data-centers by hacking support system suites.
http://magazine.hitb.org/issues/HITB-Ezine-Issue-004.pdf
Open Redirect Wreck Off - Differential Redirection Attacks
http://magazine.hitb.org/issues/HITB-Ezine-Issue-002.pdf
@Soroush: thank you, added #53.
@SecNiche: thanks for the contribution. I added 3 of the 6 you commented, #54 - #56. The others, while interesting articles, did not appear to be new techniques. More using older techniques, while still valid, to attack more modern systems. Good luck!
@Jer Thanks.
Poisoning proxy caches using Java/Flash/Web Sockets:
http://www.adambarth.com/experimental/websocket.pdf
http://news.cnet.com/8301-30685_3-20025272-264.html
Some more for your consideration:
http://blog.andlabs.org/2010/12/performing-ddos-attacks-with-html5.html
http://blog.andlabs.org/2010/12/cracking-hashes-in-javascript-cloud.html
http://blog.andlabs.org/2010/07/shell-of-future-reverse-web-shell.html
Some more for your consideration:
http://blog.andlabs.org/2010/12/cracking-hashes-in-javascript-cloud.html
http://blog.andlabs.org/2010/12/performing-ddos-attacks-with-html5.html
http://blog.andlabs.org/2010/07/shell-of-future-reverse-web-shell.html
@Collin: Thanks! Got you down at #62
@Lava: Added 2 of your three. Excellent work.
Plus I added two more.
Awesome, thanks :)
Would my phishing variant qualify?
http://skeletonscribe.blogspot.com/2010/12/chronofeit-phishing.html
Two more suggestions:
1) Getting the cookies using a server-side redirect and JAVA by LeverOne:
http://sla.ckers.org/forum/read.php?2,35422
2) d0z.me by Ben Schmidt: http://spareclockcycles.org/2010/12/19/d0z-me-the-evil-url-shortener/
@albino: thank you, #65
@lava: thanks, but this is too similar to earlier work:
http://jeremiahgrossman.blogspot.com/2008/04/csrf-ddos-skeleton-in-closet.html
The IIS 5.1 one using alternate data streams is a very, very old attack, but a new form. Appending :$DATA to filenames has been known for at least a decade.
While this is very specific to the Django web framework. It's essentially blind injection but at the ORM level.
http://evilpacket.net/2010/dec/22/information-leakage-django-administrative-interfac/
SQL injection filter evasion by Reiners:
http://websec.files.wordpress.com/2010/11/sqli2.pdf (slides)
https://websec.wordpress.com/2010/12/04/sqli-filter-evasion-cheat-sheet-mysql/ (cheatsheet)
Interesting one if it's not too late:
http://briandefrancesco.com/?p=40
Probably it's too late but still good to be in the list for the future reference:
-
Breaking HTML parsers for fun
http://www.thespanner.co.uk/2010/11/25/breaking-html-parsers-for-fun/
-
setTimeout and setInterval
http://www.thespanner.co.uk/2010/09/10/settimeout-and-setinterval/
-
JSReg bypasses:
http://www.thespanner.co.uk/2010/10/31/jsreg-bypasses/
http://code.google.com/p/jsreg/wiki/Exploits
http://rgaucher.info/planet/The_Spanner/2010/11/07/Soroush_Dalili_breaks_JSReg_again
-
x5s - test encodings and character transformations to find XSS hotspots
http://xss.codeplex.com/
http://www.lookout.net/2010/12/20/list-of-characters-for-testing-unicode-transformations-and-best-fit-mapping-to-dangerous-ascii/
-
Facebook Redirect Link – New Bypass Method – “:/” after the domain name
http://soroush.secproject.com/blog/2010/12/facebook-redirect-link-new-bypass-method-%E2%80%93-%E2%80%9C%E2%80%9D-after-the-domain-name/
-
;)
How come the best server side findings and techniques (Struts/JBoss/Spring) from Meder didn't make into the list? . Also, where is the Java Trusted Method Chaining by Sami Koivu?
You might argue with the "no CVE rule" but there is no clear cut between techniques and bug in these cases actually. For example, in order to have a successful attack on ASP.NET, POET need to exploit bug of ASP.NET implementation/configuration but you have POET on the list]
@zdx: "making the list" is subject to me finding it through my personal efforts or people submitting them on their own, with some light validation of course. That's a big reason for the effort, to capture everything that's been learned over the past year and not have it get lost in the ether as has been the problem in years past.
If you can supply the best reference links to the attacks you mentioned, I'd be very happy to review them for inclusion on the big list. The CVE mention was just an indication that we want "new" technique, not individual bug instances.
I like to add a vulnerability which I found during a gray box testing..
Its a type of back refresh attack. For mitigating back refresh attack we use a 302 redirect in successful operation. But in one of my app i tested there was a password policy saying password cant change withing 30 days. attack is If the user try to change the password withing 30 days it will respond with a 200 ok response. For a successful change adversary cant do anything..
But if the password changing failed due to some reason(password policy, new & conform password mismatch.. etc)attacker can exploit the back refresh option of the browser and can capture the request in proxy.
New google hacking techniques(Type the commands in google search engine.):
1. accessing public security cameras-
inul:"viewframe?mode=motion"
intitle:"Live View / - AXIS"
2. Un spider sites:
"robots.txt" "disallow:" filetype:txt
3. Front page user info-
inurl:_vti_pvt "service.pwd"
4. PHP Photo album-
inurl:"phphotoalbum/upload"
5. VNC user info-
"vnc desktop" inurl:5800
6. Network printer-
inurl:"port_255" -htm
7.PHP admin account-
intitle:phpMyAdmin "Welcome to phpMyAdmin***" "running on*as root@*"
I would like to propose my seminar work "UI Redressing: Attacks and Countermeasures Revisted".
URL: http://ui-redressing.mniemietz.de/
Thanks in advance.
@shinto143: The back button attack you describe, cool as it is, has actually been documented and demoed in years past. Don't have time to find the reference at the moment. And your Google Hacking concept has been around for a while, I don't see any "new techniques" described.
@Marcus Niemietz: thanks for the submission. I can certainly add it to the big list, but can't get it voted on now since the process has commenced.
You guys shouldn't be posting advice to hack.
Post a Comment