SlideShare a Scribd company logo
Thug: a low-interaction honeyclient
Angelo Dell'Aera
Speaker

Chief Executive Officer @ Honeynet Project

Information Security Independent Researcher @ Antifork
Research (10+ years)
Agenda

Introduction

Honeyclient technologies

Thug

Conclusions
Client-side attacks

The number of client-side attacks has grown significantly in
the past few years shifting focus on poorly protected
vulnerable clients

In the last years more and more attacks against client
systems

The browser is the most popular client system deployed on
every user system

A lot of vulnerabilities are identified daily and reported in the
most used browsers and in third-party plugins
Honeyclients

Just as the most known honeypot technologies enable
research into server-side attacks, honeyclients allow the
study of client-side attacks

A complement to honeypots, a honeyclient is a tool
designed to mimic the behavior of a user-driven network
client application, such as a web browser, and be
exploited by an attacker’s content
Honeyclients

What we need is something which seems like a real
browser the same way as a classical honeypot system
seems like a real vulnerable server

A real system (high-interaction honeyclient) or an
emulated one (low-interaction honeyclient)?
Document Object Model (DOM)
“The Document Object Model is a platform- and language-neutral interface that
will allow programs and scripts to dynamically access and update the content,
structure and style of documents. The document can be further processed and the
results of that processing can be incorporated back into the presented page.”

Thug DOM is (almost) compliant with W3C DOM Core,
HTML, Events and Views specifications (Level 1, 2 and
partially 3) and partially compliant with W3C DOM Style
specifications

Designed with the requirement that adding the missing
features has to be as simple as possible
Document Object Model (DOM)
Browser Personalities

Drive-by download attacks target specific versions of the
browser so a properly designed low-interaction honeyclient
should be able to emulate different browser personalities

Supporting different browser personalities is almost a
matter of implementing different (and sometimes totally
incompatible) DOM behaviors and interfaces
Supported Browser Personalities – 1/2
➢
Internet Explorer 6.0 (Windows XP)
➢
Internet Explorer 6.1 (Windows XP)
➢
Internet Explorer 7.0 (Windows XP)
➢
Internet Explorer 8.0 (Windows XP)
➢
Chrome 20.0.1132.47 (Windows XP)
➢
Firefox 12.0 (Windows XP)
➢
Safari 5.1.7 (Windows XP)
➢
Internet Explorer 6.0 (Windows 2000)
➢
Internet Explorer 8.0 (Windows 2000)
➢
Internet Explorer 8.0 (Windows 7)
➢
Internet Explorer 9.0 (Windows 7)
➢
Chrome 20.0.1132.47 (Windows 7)
➢
Firefox 3.6.13 (Windows 7)
➢
Safari 5.1.7 (Windows 7)
Supported Browser Personalities – 2/2
➢
Safari 5.1.1 (MacOS X 10.7.2)
➢
Chrome 19.0.1084.54 (MacOS X 10.7.4)
➢
Chrome 26.0.1410.19 (Linux)
➢
Chrome 30.0.1599.15 (Linux)
➢
Firefox 19.0 (Linux)
➢
Chrome 18.0.1025.166 (Samsung Galaxy S II, Android 4.0.3)
➢
Chrome 25.0.1364.123 (Samsung Galaxy S II, Android 4.0.3)
➢
Chrome 29.0.1547.59 (Samsung Galaxy S II, Android 4.1.2)
➢
Chrome 18.0.1025.133 (Google Nexus, Android 4.0.4)
➢
Safari 7.0 (iPad, iOS 7.0.4)
Document Object Model (DOM)
Event Handling
•
W3C DOM Events specification constitute the most
difficult one to emulate because of the (sometimes huge)
differences in how different browsers handle events
•
Thug emulates the different behaviors of the supported
browsers emulating load and mousemove events by
default and allowing to emulate all the other ones if
needed
Document Object Model (DOM)
Hooks
•
Thug defines some DOM hooks which are useful for
analyzing well-known exploits
•
The next example shows how Thug implements an hook
for analyzing a Java exploit with security prompt/warning
bypass (CVE-2013-2423)
Document Object Model (DOM)
Hooks
def _handle_jnlp(self, data, headers):
try:
soup = BeautifulSoup.BeautifulSoup(data)
except:
return
if soup.find("jnlp") is None:
return
log.ThugLogging.add_behavior_warn(description = '[JNLP Detected]', method = 'Dynamic Analysis')
for param in soup.find_all('param'):
log.ThugLogging.add_behavior_warn(description = '[JNLP] %s' % (param, ),
method = 'Dynamic Analysis')
self._check_jnlp_param(param)
jar = soup.find("jar")
if jar is None:
return
try:
url = jar.attrs['href']
headers['User-Agent'] = self.javaWebStartUserAgent
response, content = self.window._navigator.fetch(url, headers = headers, redirect_type = "JNLP")
except:
pass
Javascript

Google V8 Javascript engine wrapped through PyV8

“V8 implements ECMAScript as specified in ECMA-262, 5th edition,
and runs on Windows, Mac OS X , and Linux systems that use IA-32,
x64, or ARM processors. The V8 API provides functions for compiling
and executing scripts, accessing C++ methods and data structures,
handling errors, and enabling security checks”
•
Abstract Syntax Tree generation and inspection (static
analysis)
•
Context inspection (dynamic analysis)
•
Other potentially interesting features (GDB JIT interface,
live objects inspection, code disassembler, etc.) exported
through a clean and well designed API
Analysis

Static analysis
➢
Abstract Syntax Tree (AST)

Dynamic analysis

V8 debugger protocol

Libemu integration (shellcode detection and
emulation)
Abstract Syntax Tree (AST)
•
Static analysis
➢
Static attack signatures
➢
Interesting breakpoints identification for later dynamic
analysis
➢
Symbols identification for later dynamic analysis
•
Easily built through V8 API
•
Thug AST implementation is quite generic and extensible
and allows easily building and inspecting the tree
Vulnerability Modules

Python-based vulnerability modules

ActiveX controls

Core browser functionalities

Browser plugins
ActiveX
•
Thug implements an ActiveX layer of its own for emulating
ActiveX controls (just for Internet Explorer personalities)
•
It makes use of (Python) vulnerability modules in order to
emulate such ActiveX controls or just some of their
methods and attributes
•
The layer was designed in order to allow adding new
ActiveX controls in a fast and easy way
Browser Plugins
•
Drive-by download attacks target specific versions of the
browser plugins so a properly designed low-interaction
honeyclient should be able to emulate different browser
plugins versions or to disable them
-A, --adobepdf= Specify the Adobe Acrobat Reader version (default: 9.1.0)
-P, --no-adobepdf Disable Adobe Acrobat Reader plugin
-S, --shockwave= Specify the Shockwave Flash version (default: 10.0.64.0)
-R, --no-shockwave Disable Shockwave Flash plugin
-J, --javaplugin= Specify the JavaPlugin version (default: 1.6.0.32)
-K, --no-javaplugin Disable Java plugin
Logging
•
MITRE MAEC logging format
•
JSON logging format (contributed by Avira)
•
Exploit graph (contributed by Avira)
•
“Flat” log files (not so exciting I know)
•
MongoDB
•
HPFeeds

thug.events channel (URL analysis results published in
MAEC format)

thug.files channel (downloaded samples)
Logging
Blackhole 2.0 Exploit Kit
{
“timestamp”: “2013-04-13 13:43:54.307237”,
“cve”: “None”,
"description": [window open redirection] about:blank->
hxxp://purrfectpetresort.com/news/wanting_book_switch.php“,
"method": "Dynamic Analysis"
}
Honeynet Project Workshop 2014 - Thug: a low-interaction honeyclient
Logging
{
"mimetype": "PE32 executable (GUI) Intel 80386 (stripped to external PDB), for MS Windows",
"url": "hxxp://cadgrad.com/adobe/update_flash_player.exe",
"flags": {},
"sha256": "d59d9af4e9ec25431acfd8938895b5c3b728d818db024d76f5aa265e0b171f4f",
"content-type": "application/octet-stream",
"md5": "a3266663f644dc0c0df42e8da1404878",
"size": 130560
}
•
Classifiers support was introduced in Thug 0.4.24 and is
based on Yara signatures
•
Currently two classifiers exist:
➢
URL classifier
➢
Javascript classifier
Classifiers
The URL classifier works on URL pattern matching trying to
identify typical exploit kits URL i.e.
rule Blackhole_V2_2 : Exploit_Kit {
meta:
author = "Thorsten Sick"
strings:
$url = //closest/w{15,35}.php/ nocase
condition:
$url
}
URL Classifier
The Javascript classifier exploits the idea that even if the code is
obfuscated Thug goes through all the deobfuscation stages.
Working this way it can catch details which does not change so
frequently in a typical exploit kit i.e.
rule PluginDetect : Multiple_Exploit_Kits {
meta:
author = "Angelo Dell'Aera"
strings:
$jar = "getjavainfo.jar" nocase
$pdpd = "pdpd" nocase
$getver = "getversion" nocase
condition:
($jar or $pdpd) and $getver
}
Javascript Classifier
Thug source code is publicly available at
https://github.com/buffer/thug
Contributions, comments and feedback welcome!
Source code
Thanks for the attention!
Questions?
Angelo Dell'Aera
<angelo.dellaera@honeynet.org>

More Related Content

KEY
PyCon AU 2010 - Getting Started With Apache/mod_wsgi.
PDF
Java Security Manager Reloaded - jOpenSpace Lightning Talk
KEY
PyCon US 2012 - State of WSGI 2
PDF
Java Security Manager Reloaded - Devoxx 2014
PPTX
Angular js security
PDF
[UniteKorea2013] Protecting your Android content
PDF
Puppet at janrain
PPTX
Nginx warhead
PyCon AU 2010 - Getting Started With Apache/mod_wsgi.
Java Security Manager Reloaded - jOpenSpace Lightning Talk
PyCon US 2012 - State of WSGI 2
Java Security Manager Reloaded - Devoxx 2014
Angular js security
[UniteKorea2013] Protecting your Android content
Puppet at janrain
Nginx warhead

What's hot (20)

PDF
Building A Poor man’s Fir3Ey3 Mail Scanner
PPTX
Here Be Dragons: The Unexplored Land of Active Directory ACLs
PDF
HTTP For the Good or the Bad - FSEC Edition
PPTX
It's 10pm: Do You Know Where Your Writes Are?
PDF
Scala Frustrations
PDF
Windows attacks - AT is the new black
PPTX
Security in PHP - 那些在滲透測試的小技巧
PDF
Django の認証処理実装パターン / Django Authentication Patterns
PDF
Threat stack aws
PDF
Djangoアプリのデプロイに関するプラクティス / Deploy django application
PDF
Think Like a Hacker - Database Attack Vectors
PPTX
DevOps Fest 2019. Сергей Марченко. Terraform: a novel about modules, provider...
PPTX
Ansible presentation
PPTX
Devoops: DoJ Annual Cybersecurity Training Symposium Edition 2015
PDF
ElasticSearch
PDF
Windows Attacks AT is the new black
PDF
Python RESTful webservices with Python: Flask and Django solutions
PPTX
關於SQL Injection的那些奇技淫巧
PDF
soft-shake.ch - Hands on Node.js
PDF
[2B1]검색엔진의 패러다임 전환
Building A Poor man’s Fir3Ey3 Mail Scanner
Here Be Dragons: The Unexplored Land of Active Directory ACLs
HTTP For the Good or the Bad - FSEC Edition
It's 10pm: Do You Know Where Your Writes Are?
Scala Frustrations
Windows attacks - AT is the new black
Security in PHP - 那些在滲透測試的小技巧
Django の認証処理実装パターン / Django Authentication Patterns
Threat stack aws
Djangoアプリのデプロイに関するプラクティス / Deploy django application
Think Like a Hacker - Database Attack Vectors
DevOps Fest 2019. Сергей Марченко. Terraform: a novel about modules, provider...
Ansible presentation
Devoops: DoJ Annual Cybersecurity Training Symposium Edition 2015
ElasticSearch
Windows Attacks AT is the new black
Python RESTful webservices with Python: Flask and Django solutions
關於SQL Injection的那些奇技淫巧
soft-shake.ch - Hands on Node.js
[2B1]검색엔진의 패러다임 전환
Ad

Similar to Honeynet Project Workshop 2014 - Thug: a low-interaction honeyclient (20)

PDF
Eurosec2014 - An introduction to honeyclient technologies
PDF
Thug: a new low-interaction honeyclient
PPT
(In)Security Implication in the JS Universe
PDF
django
PDF
[convergese] Adaptive Images in Responsive Web Design
PDF
Oleh Zasadnyy "Progressive Web Apps: line between web and native apps become ...
ODP
Introduce Django
PDF
بررسی چارچوب جنگو
PPTX
OWASP ZAP Workshop for QA Testers
ODP
Dynamic Languages Web Frameworks Indicthreads 2009
PPT
Intelligent adware blocker symantec
PPTX
End-to-end testing with geb
PPTX
React django
PDF
Advanced iOS Debbuging (Reloaded)
ODP
Knolx session
PPTX
Raising ux bar with offline first design
PDF
Google Web Toolkit
PDF
Hack Like It's 2013 (The Workshop)
PDF
XPages Blast - ILUG 2010
Eurosec2014 - An introduction to honeyclient technologies
Thug: a new low-interaction honeyclient
(In)Security Implication in the JS Universe
django
[convergese] Adaptive Images in Responsive Web Design
Oleh Zasadnyy "Progressive Web Apps: line between web and native apps become ...
Introduce Django
بررسی چارچوب جنگو
OWASP ZAP Workshop for QA Testers
Dynamic Languages Web Frameworks Indicthreads 2009
Intelligent adware blocker symantec
End-to-end testing with geb
React django
Advanced iOS Debbuging (Reloaded)
Knolx session
Raising ux bar with offline first design
Google Web Toolkit
Hack Like It's 2013 (The Workshop)
XPages Blast - ILUG 2010
Ad

Recently uploaded (20)

PDF
Electronic commerce courselecture one. Pdf
PDF
Architecting across the Boundaries of two Complex Domains - Healthcare & Tech...
PDF
Machine learning based COVID-19 study performance prediction
PDF
Accuracy of neural networks in brain wave diagnosis of schizophrenia
PPTX
Spectroscopy.pptx food analysis technology
PPTX
Programs and apps: productivity, graphics, security and other tools
PPT
“AI and Expert System Decision Support & Business Intelligence Systems”
PDF
Diabetes mellitus diagnosis method based random forest with bat algorithm
PDF
The Rise and Fall of 3GPP – Time for a Sabbatical?
PPTX
1. Introduction to Computer Programming.pptx
PDF
Unlocking AI with Model Context Protocol (MCP)
PPTX
A Presentation on Artificial Intelligence
PDF
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
PPTX
Machine Learning_overview_presentation.pptx
PDF
cuic standard and advanced reporting.pdf
PPTX
Group 1 Presentation -Planning and Decision Making .pptx
PPTX
SOPHOS-XG Firewall Administrator PPT.pptx
PPTX
MYSQL Presentation for SQL database connectivity
PDF
Building Integrated photovoltaic BIPV_UPV.pdf
PPTX
KOM of Painting work and Equipment Insulation REV00 update 25-dec.pptx
Electronic commerce courselecture one. Pdf
Architecting across the Boundaries of two Complex Domains - Healthcare & Tech...
Machine learning based COVID-19 study performance prediction
Accuracy of neural networks in brain wave diagnosis of schizophrenia
Spectroscopy.pptx food analysis technology
Programs and apps: productivity, graphics, security and other tools
“AI and Expert System Decision Support & Business Intelligence Systems”
Diabetes mellitus diagnosis method based random forest with bat algorithm
The Rise and Fall of 3GPP – Time for a Sabbatical?
1. Introduction to Computer Programming.pptx
Unlocking AI with Model Context Protocol (MCP)
A Presentation on Artificial Intelligence
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
Machine Learning_overview_presentation.pptx
cuic standard and advanced reporting.pdf
Group 1 Presentation -Planning and Decision Making .pptx
SOPHOS-XG Firewall Administrator PPT.pptx
MYSQL Presentation for SQL database connectivity
Building Integrated photovoltaic BIPV_UPV.pdf
KOM of Painting work and Equipment Insulation REV00 update 25-dec.pptx

Honeynet Project Workshop 2014 - Thug: a low-interaction honeyclient

  • 1. Thug: a low-interaction honeyclient Angelo Dell'Aera
  • 2. Speaker  Chief Executive Officer @ Honeynet Project  Information Security Independent Researcher @ Antifork Research (10+ years)
  • 4. Client-side attacks  The number of client-side attacks has grown significantly in the past few years shifting focus on poorly protected vulnerable clients  In the last years more and more attacks against client systems  The browser is the most popular client system deployed on every user system  A lot of vulnerabilities are identified daily and reported in the most used browsers and in third-party plugins
  • 5. Honeyclients  Just as the most known honeypot technologies enable research into server-side attacks, honeyclients allow the study of client-side attacks  A complement to honeypots, a honeyclient is a tool designed to mimic the behavior of a user-driven network client application, such as a web browser, and be exploited by an attacker’s content
  • 6. Honeyclients  What we need is something which seems like a real browser the same way as a classical honeypot system seems like a real vulnerable server  A real system (high-interaction honeyclient) or an emulated one (low-interaction honeyclient)?
  • 7. Document Object Model (DOM) “The Document Object Model is a platform- and language-neutral interface that will allow programs and scripts to dynamically access and update the content, structure and style of documents. The document can be further processed and the results of that processing can be incorporated back into the presented page.”  Thug DOM is (almost) compliant with W3C DOM Core, HTML, Events and Views specifications (Level 1, 2 and partially 3) and partially compliant with W3C DOM Style specifications  Designed with the requirement that adding the missing features has to be as simple as possible
  • 8. Document Object Model (DOM) Browser Personalities  Drive-by download attacks target specific versions of the browser so a properly designed low-interaction honeyclient should be able to emulate different browser personalities  Supporting different browser personalities is almost a matter of implementing different (and sometimes totally incompatible) DOM behaviors and interfaces
  • 9. Supported Browser Personalities – 1/2 ➢ Internet Explorer 6.0 (Windows XP) ➢ Internet Explorer 6.1 (Windows XP) ➢ Internet Explorer 7.0 (Windows XP) ➢ Internet Explorer 8.0 (Windows XP) ➢ Chrome 20.0.1132.47 (Windows XP) ➢ Firefox 12.0 (Windows XP) ➢ Safari 5.1.7 (Windows XP) ➢ Internet Explorer 6.0 (Windows 2000) ➢ Internet Explorer 8.0 (Windows 2000) ➢ Internet Explorer 8.0 (Windows 7) ➢ Internet Explorer 9.0 (Windows 7) ➢ Chrome 20.0.1132.47 (Windows 7) ➢ Firefox 3.6.13 (Windows 7) ➢ Safari 5.1.7 (Windows 7)
  • 10. Supported Browser Personalities – 2/2 ➢ Safari 5.1.1 (MacOS X 10.7.2) ➢ Chrome 19.0.1084.54 (MacOS X 10.7.4) ➢ Chrome 26.0.1410.19 (Linux) ➢ Chrome 30.0.1599.15 (Linux) ➢ Firefox 19.0 (Linux) ➢ Chrome 18.0.1025.166 (Samsung Galaxy S II, Android 4.0.3) ➢ Chrome 25.0.1364.123 (Samsung Galaxy S II, Android 4.0.3) ➢ Chrome 29.0.1547.59 (Samsung Galaxy S II, Android 4.1.2) ➢ Chrome 18.0.1025.133 (Google Nexus, Android 4.0.4) ➢ Safari 7.0 (iPad, iOS 7.0.4)
  • 11. Document Object Model (DOM) Event Handling • W3C DOM Events specification constitute the most difficult one to emulate because of the (sometimes huge) differences in how different browsers handle events • Thug emulates the different behaviors of the supported browsers emulating load and mousemove events by default and allowing to emulate all the other ones if needed
  • 12. Document Object Model (DOM) Hooks • Thug defines some DOM hooks which are useful for analyzing well-known exploits • The next example shows how Thug implements an hook for analyzing a Java exploit with security prompt/warning bypass (CVE-2013-2423)
  • 13. Document Object Model (DOM) Hooks def _handle_jnlp(self, data, headers): try: soup = BeautifulSoup.BeautifulSoup(data) except: return if soup.find("jnlp") is None: return log.ThugLogging.add_behavior_warn(description = '[JNLP Detected]', method = 'Dynamic Analysis') for param in soup.find_all('param'): log.ThugLogging.add_behavior_warn(description = '[JNLP] %s' % (param, ), method = 'Dynamic Analysis') self._check_jnlp_param(param) jar = soup.find("jar") if jar is None: return try: url = jar.attrs['href'] headers['User-Agent'] = self.javaWebStartUserAgent response, content = self.window._navigator.fetch(url, headers = headers, redirect_type = "JNLP") except: pass
  • 14. Javascript  Google V8 Javascript engine wrapped through PyV8  “V8 implements ECMAScript as specified in ECMA-262, 5th edition, and runs on Windows, Mac OS X , and Linux systems that use IA-32, x64, or ARM processors. The V8 API provides functions for compiling and executing scripts, accessing C++ methods and data structures, handling errors, and enabling security checks” • Abstract Syntax Tree generation and inspection (static analysis) • Context inspection (dynamic analysis) • Other potentially interesting features (GDB JIT interface, live objects inspection, code disassembler, etc.) exported through a clean and well designed API
  • 15. Analysis  Static analysis ➢ Abstract Syntax Tree (AST)  Dynamic analysis  V8 debugger protocol  Libemu integration (shellcode detection and emulation)
  • 16. Abstract Syntax Tree (AST) • Static analysis ➢ Static attack signatures ➢ Interesting breakpoints identification for later dynamic analysis ➢ Symbols identification for later dynamic analysis • Easily built through V8 API • Thug AST implementation is quite generic and extensible and allows easily building and inspecting the tree
  • 17. Vulnerability Modules  Python-based vulnerability modules  ActiveX controls  Core browser functionalities  Browser plugins
  • 18. ActiveX • Thug implements an ActiveX layer of its own for emulating ActiveX controls (just for Internet Explorer personalities) • It makes use of (Python) vulnerability modules in order to emulate such ActiveX controls or just some of their methods and attributes • The layer was designed in order to allow adding new ActiveX controls in a fast and easy way
  • 19. Browser Plugins • Drive-by download attacks target specific versions of the browser plugins so a properly designed low-interaction honeyclient should be able to emulate different browser plugins versions or to disable them -A, --adobepdf= Specify the Adobe Acrobat Reader version (default: 9.1.0) -P, --no-adobepdf Disable Adobe Acrobat Reader plugin -S, --shockwave= Specify the Shockwave Flash version (default: 10.0.64.0) -R, --no-shockwave Disable Shockwave Flash plugin -J, --javaplugin= Specify the JavaPlugin version (default: 1.6.0.32) -K, --no-javaplugin Disable Java plugin
  • 20. Logging • MITRE MAEC logging format • JSON logging format (contributed by Avira) • Exploit graph (contributed by Avira) • “Flat” log files (not so exciting I know) • MongoDB • HPFeeds  thug.events channel (URL analysis results published in MAEC format)  thug.files channel (downloaded samples)
  • 21. Logging Blackhole 2.0 Exploit Kit { “timestamp”: “2013-04-13 13:43:54.307237”, “cve”: “None”, "description": [window open redirection] about:blank-> hxxp://purrfectpetresort.com/news/wanting_book_switch.php“, "method": "Dynamic Analysis" }
  • 23. Logging { "mimetype": "PE32 executable (GUI) Intel 80386 (stripped to external PDB), for MS Windows", "url": "hxxp://cadgrad.com/adobe/update_flash_player.exe", "flags": {}, "sha256": "d59d9af4e9ec25431acfd8938895b5c3b728d818db024d76f5aa265e0b171f4f", "content-type": "application/octet-stream", "md5": "a3266663f644dc0c0df42e8da1404878", "size": 130560 }
  • 24. • Classifiers support was introduced in Thug 0.4.24 and is based on Yara signatures • Currently two classifiers exist: ➢ URL classifier ➢ Javascript classifier Classifiers
  • 25. The URL classifier works on URL pattern matching trying to identify typical exploit kits URL i.e. rule Blackhole_V2_2 : Exploit_Kit { meta: author = "Thorsten Sick" strings: $url = //closest/w{15,35}.php/ nocase condition: $url } URL Classifier
  • 26. The Javascript classifier exploits the idea that even if the code is obfuscated Thug goes through all the deobfuscation stages. Working this way it can catch details which does not change so frequently in a typical exploit kit i.e. rule PluginDetect : Multiple_Exploit_Kits { meta: author = "Angelo Dell'Aera" strings: $jar = "getjavainfo.jar" nocase $pdpd = "pdpd" nocase $getver = "getversion" nocase condition: ($jar or $pdpd) and $getver } Javascript Classifier
  • 27. Thug source code is publicly available at https://github.com/buffer/thug Contributions, comments and feedback welcome! Source code
  • 28. Thanks for the attention! Questions? Angelo Dell'Aera <angelo.dellaera@honeynet.org>