Skip to content

dirtyfilthy/lulz

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

67 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Lulz is distributed under the http://creativecommons.org/licenses/by-nc-sa/3.0/ 
licence.

THIS IS VERY ALPHA SOFTWARE, lots of bugs, bad code, etc. Feedback, bug reports 
and suggestions gratefully accepted at [email protected]

More information, latest releases etc available at 
http://www.dirtyfilthy.net/projects/lulz

#####################
## Introduction
#########################

Lulz is a webstalking engine & command-line tool for fast online personal 
reconnaissance: searching, collating and analyzing public online profiles.

Lulz is a webstalking engine that tries to search for information about a 
particular person online in a similar fashion to the way a human being would do 
it, that is, by finding, evaluating and utilising information. When you 
webstalk someone you typically take what you know and try to transform this 
knowledge into more data, using a variety of methods (such as search engines) 
and tricks (such as social network account enumeration via email). You then 
evaluate the information (is this who I'm looking for?) and if so lather, 
rinse, repeat. 

Lulz uses Bayesian identity resolution to determine if person A and person B 
are the same person, an agent architecture to utilise the information it finds, 
and a blackboard architecture to share information amongst these agents. Lulz 
uses a Markov Decision Process to decide what to do next. 

#######################
## Usage
##########################

For usage info ./lulz -h

At the moment lulz is very untrained, i recommend using a timeout, i.e. 
--timeout 420. The timeout only prevents lulz from running new agents, it will 
still finish transforming and processing the profile queue. Use --hard-timeout 
to stop all processing at a certain point. Also the output can be quite long, 
might be an idea to redirect output to a file. You can slim this output down by 
setting a high match value, say -m 0.8

The more information you enter at the start, the better lulz works, but also: 
garbage in, garbage out. Entering incorrect information will definitely reduce 
effectiveness and may easily produce nonsense. Keywords sometimes help greatly 
in distinguishing between profiles, i.e. if your target works in the computer 
security industry use --keyword computer --keyword security
 
#######################
## Example search      
##########################

./lulz --alias dirtyfilthy --sex m --country nz --city christchurch --homepage 
www.dirtyfilthy.net --summary -m 0.8 --timeout 240 --hard-timeout 500

#######################
## Analysis Examples
##########################

./lulz --analyze twitter.com/dirtyfilthy

Analyze this twitter account

./lulz -r --analyze twitter.com/dirtyfilthy

Analyze this twitter account & dump posts (archive)

./lulz -X -r --analyze twitter.com/dirtyfilthy

Analyze this twitter account & dump everything in xml

#######################
## Status key
##########################

.	- running transformer
=	- running parser
*	- running search
?  	- performing identity resolution
!  	- recalculating action queue
W  	- queue empty, waiting
F  	- maximum number of agent threads reached
T  	- timeout reached, waiting on identity resolution & transformers to 
finish
A	- running analysis

Configuration & resources are setup in yaml files in config/defaults.yml and 
config/resource.yml respectively.

Releases

No releases published

Packages

No packages published