-
Notifications
You must be signed in to change notification settings - Fork 516
/
cs-extract-key.py
1795 lines (1629 loc) · 72.5 KB
/
cs-extract-key.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#!/usr/bin/env python
from __future__ import print_function
__description__ = 'Extract cryptographic keys from Cobalt Strike beacon process dump'
__author__ = 'Didier Stevens'
__version__ = '0.0.4'
__date__ = '2021/12/11'
"""
Source code put in the public domain by Didier Stevens, no Copyright
https://DidierStevens.com
Use at your own risk
History:
2021/04/19: start
2021/04/21: version 3 and 4
2021/04/22: codepages
2021/04/25: option -c
2021/10/07: 0.0.2 updated missing modules logic
2021/10/17: made some key search space improvements
2021/10/28: added option -f
2021/10/29: added option -t
2021/10/31: changes to output
2021/11/02: man page
2021/11/06: 0.0.3 added AverageDifferenceConsecutiveBytes and disabled it
2021/11/11: added option verbose
2021/12/11: 0.0.4 added option donotfullsearch
Todo:
Document flag arguments in man page
"""
import optparse
import sys
import os
import zipfile
import binascii
import random
import gzip
import collections
import glob
import textwrap
import re
import struct
import string
import math
import fnmatch
import json
import time
import csv
import hashlib
import hmac
try:
import Crypto.Cipher.AES
except ImportError:
print('Crypto.Cipher.AES module required: pip install pycryptodome')
exit(-1)
if sys.version_info[0] >= 3:
from io import BytesIO as DataIO
else:
from cStringIO import StringIO as DataIO
if sys.version_info[0] >= 3:
from io import StringIO
else:
from cStringIO import StringIO
def PrintManual():
manual = r'''
Manual:
This tool extracts cryptographic keys, used by Cobalt Strike beacons and team servers to encrypt their C2 communication, from the beacon's process memory.
This tool takes one or more process memory dump files as input. These files will typically be minidump (.dmp) files created with Sysinternals' tools, they can be actually any memory dump, as long as it can fit in Python's memory.
The process memory dump files are handled by this tool as flat, raw data. This tool will not parse minidump file structures (this is foreseen as a feature in another upcoming tool).
As long as the data artifacts processed by this tool are not compressed or encoded, any memory dump is OK.
These data artifacts are found in writable process memory, so a minidump with only writable process memory is sufficient (procdump's option -mp).
When provided with a process memory dump, without using any option, this tool will look for unencrypted Beacon metadata inside the process memory dump.
This metadata starts with a sequence of 4 bytes: 00 00 BE EF. This tool searches for this sequence and parses the metadata that follows it.
Here is an example:
Command: cs-extract-key.py beacon.dmp
Output:
File: beacon.dmp
Position: 0x009094d0
Header: 0000beef
Datasize: 00000045
Raw key: 61f9880e25408a44ae43cdedb642f099
aeskey: 9f71c29ed793f778189a10aae54563cb
hmackey: e236ddf9d75b440484a0550e1ca3e2e8
charset: 04e4 ANSI Latin 1; Western European (Windows)
charset_oem: 01b5 OEM United States
Field: b'65153'
Field: b'2744'
Field: b'6.1'
Field: b'127.0.0.1'
Field: b'WIN-DFGODF5ES9C'
Field: b'root'
Field: b'1'
Field: b'1'
AES key:
Position: 0x001908e0
HMAC key:
Position: 0x001908f0
Right after the header (0x0000BEEF) is an integer with the datasize of the metadata. This should not be too long, very rarely will it be longer than 200 bytes. If it is 1000 or more bytes, then this is a clear indication that this is a false positive: that the found sequence 0x0000BEEF is not the start of a metadata byte sequence.
The "raw key" decoded from the metadata, is a random sequence of 16 bytes generated by the beacon (unique for each process) that serves to generate the HMAC and AES keys used by the beacon and team server to encrypt their communication.
The algorithm is: calculate the SHA256 hash of the raw key, take the first half of the SHA256 as the HMAC key, and take the second half as the AES key.
Thus the HMAC key and AES key itself are not contained inside the metadata, but are calculated from the raw key that is inside the metadata.
Several fields are found after the raw key, like the computername and username running the beacon. These fields can be used to validate that valid metadata was found. If these fields don't look like expected computernames and usernames, then we are most likely dealing with a false positive, that is best ignored.
After deriving the HMAC and AES key from the raw key, this tool will try to find these 2 keys in process memory. These are the Position: entries found in the output example above. If these HMAC and AES keys are not found inside process memory, then we are also most likely dealing with a false positive.
In our experience, detection of metadata as explained above, is only successful with Cobalt Strike version 3.x beacons. And preferably with process memory dumps taken early in the lifespan of a running beacon.
To extract cryptographic keys from process memory of Cobalt Strike version 4.x beacons, another method must be followed.
This second method requires encrypted data obtained from a network capture file (extracted manually or with tool cs-parse-http-traffic.py).
Unlike metadata, HMAC and AES keys have no features that distinguishes them from other data in process memory. They are just 16 bytes of data that looks random.
To detect HMAC and AES keys inside process memory, the tool will proceed with a kind of dictionary attack: all possible 16-byte long, non-null byte sequences obtained from process memory, are considered as potential keys and used in attempts to decrypt and validate the encrypted data.
If this succeeds, a valid key was found.
I have observed that the HMAC and AES keys are often found in process memory, somewhere after string sha256\x00. As a speed optimization tactic, this tool will search for each occurrence of string sha256\x00 and try out the 0x500000 bytes that follow these string occurrences as HMAC and AES keys.
For a typical writable process memory dumb of no more than 10MB, this takes a couple of minutes.
If no occurrences of string sha250\x00 are found, then the complete process memory dump is processed, unless option -d is used. This full search processing mode can be forced with option -f.
There is a small difference between the way that data is encrypted by the beacon and the way that data is encrypted by the team server.
Encrypted data sent by the team server to the beacon, contains tasks to be executed by the beacon. That encrypted data looks completely random. Option -t (task) must be used to provide this encrypted data (as a hexadecimal string) to the tool.
Encrypted data sent by the beacon to the team server, contains output from the tasks executed by the beacon. This output is called a callback. That encrypted data looks almost completely random. The first 4 bytes of that data represent the length of the encrypted data, encoded as an unsigned, 32-bit big-endian integer. Thus the first bytes of a callback are often 00 bytes. Callbacks can be concatenated together in one POST (default) request. Option -c (callback) must be used to provide this encrypted data (as a hexadecimal string) to the tool.
Here is an example:
Command: cs-extract-key.py -t d12c14aa698a6b85a8ed3c3c33774fe79acadd0e95fa88f45b66d8751682db734472b2c9c874ccc70afa426fb2f510654df7042aa7d2384229518f26d1e044bd beacon.dmp
Output:
File: beacon.dmp
Searching for AES and HMAC keys
Searching after sha256\x00 string (0x6a8179)
AES key position: 0x006ae6c5
AES Key: 23a79f098615fdd74fbb25116f50aa09
HMAC key position: 0x006b19e5
HMAC Key: 5a8a1e3d8c75f37353937475bd498dfb
SHA256 raw key: 5a8a1e3d8c75f37353937475bd498dfb:23a79f098615fdd74fbb25116f50aa09
Searching for raw key
d12c14aa698a6b85a8ed3c3c33774fe79acadd0e95fa88f45b66d8751682db734472b2c9c874ccc70afa426fb2f510654df7042aa7d2384229518f26d1e044bd is encrypted data (a task), obtained from a network traffic capture file, that is passed to the tool with option -t.
In the example above, the AES and HMAC key are recovered, but not the raw key (the raw key is typically not found with version 4.x beacons).
Recovered HMAC keys are 100% guaranteed to be true positives, since they validate the HMAC signature of the encrypted data.
Recovered AES keys can sometimes be false positives: the decrypted data looks like a task, but it's not actually a task.
But whenever I found an AES key near an HMAC key, they were always true positives. I only had false positives when an AES key was recovered without recovering a HMAC key.
Remark that the above method (using option -t or -c) works also for version 3.x beacons.
Beacon process memory can be encoded while the beacon is sleeping. This is done with a configuration option called a sleep mask. Since beacons sleep most of the time, it is very likely that you will take a process dump while a beacon is sleeping. This tool can not recover cryptographic keys from the process memory of a beacon with a sleep mask. If that is the case, use my tool cs-analyze-processdump.py first.
'''
for line in manual.split('\n'):
print(textwrap.fill(line, 79))
DEFAULT_SEPARATOR = ','
QUOTE = '"'
CS_FIXED_IV = b'abcdefghijklmnop'
def PrintError(*args, **kwargs):
print(*args, file=sys.stderr, **kwargs)
#Convert 2 Bytes If Python 3
def C2BIP3(string):
if sys.version_info[0] > 2:
return bytes([ord(x) for x in string])
else:
return string
#Convert 2 Integer If Python 2
def C2IIP2(data):
if sys.version_info[0] > 2:
return data
else:
return ord(data)
# CIC: Call If Callable
def CIC(expression):
if callable(expression):
return expression()
else:
return expression
# IFF: IF Function
def IFF(expression, valueTrue, valueFalse):
if expression:
return CIC(valueTrue)
else:
return CIC(valueFalse)
#-BEGINCODE cBinaryFile------------------------------------------------------------------------------
#import random
#import binascii
#import zipfile
#import gzip
#import sys
#if sys.version_info[0] >= 3:
# from io import BytesIO as DataIO
#else:
# from cStringIO import StringIO as DataIO
def LoremIpsumSentence(minimum, maximum):
words = ['lorem', 'ipsum', 'dolor', 'sit', 'amet', 'consectetur', 'adipiscing', 'elit', 'etiam', 'tortor', 'metus', 'cursus', 'sed', 'sollicitudin', 'ac', 'sagittis', 'eget', 'massa', 'praesent', 'sem', 'fermentum', 'dignissim', 'in', 'vel', 'augue', 'scelerisque', 'auctor', 'libero', 'nam', 'a', 'gravida', 'odio', 'duis', 'vestibulum', 'vulputate', 'quam', 'nec', 'cras', 'nibh', 'feugiat', 'ut', 'vitae', 'ornare', 'justo', 'orci', 'varius', 'natoque', 'penatibus', 'et', 'magnis', 'dis', 'parturient', 'montes', 'nascetur', 'ridiculus', 'mus', 'curabitur', 'nisl', 'egestas', 'urna', 'iaculis', 'lectus', 'maecenas', 'ultrices', 'velit', 'eu', 'porta', 'hac', 'habitasse', 'platea', 'dictumst', 'integer', 'id', 'commodo', 'mauris', 'interdum', 'malesuada', 'fames', 'ante', 'primis', 'faucibus', 'accumsan', 'pharetra', 'aliquam', 'nunc', 'at', 'est', 'non', 'leo', 'nulla', 'sodales', 'porttitor', 'facilisis', 'aenean', 'condimentum', 'rutrum', 'facilisi', 'tincidunt', 'laoreet', 'ultricies', 'neque', 'diam', 'euismod', 'consequat', 'tempor', 'elementum', 'lobortis', 'erat', 'ligula', 'risus', 'donec', 'phasellus', 'quisque', 'vivamus', 'pellentesque', 'tristique', 'venenatis', 'purus', 'mi', 'dictum', 'posuere', 'fringilla', 'quis', 'magna', 'pretium', 'felis', 'pulvinar', 'lacinia', 'proin', 'viverra', 'lacus', 'suscipit', 'aliquet', 'dui', 'molestie', 'dapibus', 'mollis', 'suspendisse', 'sapien', 'blandit', 'morbi', 'tellus', 'enim', 'maximus', 'semper', 'arcu', 'bibendum', 'convallis', 'hendrerit', 'imperdiet', 'finibus', 'fusce', 'congue', 'ullamcorper', 'placerat', 'nullam', 'eros', 'habitant', 'senectus', 'netus', 'turpis', 'luctus', 'volutpat', 'rhoncus', 'mattis', 'nisi', 'ex', 'tempus', 'eleifend', 'vehicula', 'class', 'aptent', 'taciti', 'sociosqu', 'ad', 'litora', 'torquent', 'per', 'conubia', 'nostra', 'inceptos', 'himenaeos']
sample = random.sample(words, random.randint(minimum, maximum))
sample[0] = sample[0].capitalize()
return ' '.join(sample) + '.'
def LoremIpsum(sentences):
return ' '.join([LoremIpsumSentence(15, 30) for i in range(sentences)])
STATE_START = 0
STATE_IDENTIFIER = 1
STATE_STRING = 2
STATE_SPECIAL_CHAR = 3
STATE_ERROR = 4
FUNCTIONNAME_REPEAT = 'repeat'
FUNCTIONNAME_RANDOM = 'random'
FUNCTIONNAME_CHR = 'chr'
FUNCTIONNAME_LOREMIPSUM = 'loremipsum'
def Tokenize(expression):
result = []
token = ''
state = STATE_START
while expression != '':
char = expression[0]
expression = expression[1:]
if char == "'":
if state == STATE_START:
state = STATE_STRING
elif state == STATE_IDENTIFIER:
result.append([STATE_IDENTIFIER, token])
state = STATE_STRING
token = ''
elif state == STATE_STRING:
result.append([STATE_STRING, token])
state = STATE_START
token = ''
elif char >= '0' and char <= '9' or char.lower() >= 'a' and char.lower() <= 'z':
if state == STATE_START:
token = char
state = STATE_IDENTIFIER
else:
token += char
elif char == ' ':
if state == STATE_IDENTIFIER:
result.append([STATE_IDENTIFIER, token])
token = ''
state = STATE_START
elif state == STATE_STRING:
token += char
else:
if state == STATE_IDENTIFIER:
result.append([STATE_IDENTIFIER, token])
token = ''
state = STATE_START
result.append([STATE_SPECIAL_CHAR, char])
elif state == STATE_STRING:
token += char
else:
result.append([STATE_SPECIAL_CHAR, char])
token = ''
if state == STATE_IDENTIFIER:
result.append([state, token])
elif state == STATE_STRING:
result = [[STATE_ERROR, 'Error: string not closed', token]]
return result
def ParseFunction(tokens):
if len(tokens) == 0:
print('Parsing error')
return None, tokens
if tokens[0][0] == STATE_STRING or tokens[0][0] == STATE_IDENTIFIER and tokens[0][1].startswith('0x'):
return [[FUNCTIONNAME_REPEAT, [[STATE_IDENTIFIER, '1'], tokens[0]]], tokens[1:]]
if tokens[0][0] != STATE_IDENTIFIER:
print('Parsing error')
return None, tokens
function = tokens[0][1]
tokens = tokens[1:]
if len(tokens) == 0:
print('Parsing error')
return None, tokens
if tokens[0][0] != STATE_SPECIAL_CHAR or tokens[0][1] != '(':
print('Parsing error')
return None, tokens
tokens = tokens[1:]
if len(tokens) == 0:
print('Parsing error')
return None, tokens
arguments = []
while True:
if tokens[0][0] != STATE_IDENTIFIER and tokens[0][0] != STATE_STRING:
print('Parsing error')
return None, tokens
arguments.append(tokens[0])
tokens = tokens[1:]
if len(tokens) == 0:
print('Parsing error')
return None, tokens
if tokens[0][0] != STATE_SPECIAL_CHAR or (tokens[0][1] != ',' and tokens[0][1] != ')'):
print('Parsing error')
return None, tokens
if tokens[0][0] == STATE_SPECIAL_CHAR and tokens[0][1] == ')':
tokens = tokens[1:]
break
tokens = tokens[1:]
if len(tokens) == 0:
print('Parsing error')
return None, tokens
return [[function, arguments], tokens]
def Parse(expression):
tokens = Tokenize(expression)
if len(tokens) == 0:
print('Parsing error')
return None
if tokens[0][0] == STATE_ERROR:
print(tokens[0][1])
print(tokens[0][2])
print(expression)
return None
functioncalls = []
while True:
functioncall, tokens = ParseFunction(tokens)
if functioncall == None:
return None
functioncalls.append(functioncall)
if len(tokens) == 0:
return functioncalls
if tokens[0][0] != STATE_SPECIAL_CHAR or tokens[0][1] != '+':
print('Parsing error')
return None
tokens = tokens[1:]
def InterpretInteger(token):
if token[0] != STATE_IDENTIFIER:
return None
try:
return int(token[1])
except:
return None
def Hex2Bytes(hexadecimal):
if len(hexadecimal) % 2 == 1:
hexadecimal = '0' + hexadecimal
try:
return binascii.a2b_hex(hexadecimal)
except:
return None
def InterpretHexInteger(token):
if token[0] != STATE_IDENTIFIER:
return None
if not token[1].startswith('0x'):
return None
bytes = Hex2Bytes(token[1][2:])
if bytes == None:
return None
integer = 0
for byte in bytes:
integer = integer * 0x100 + C2IIP2(byte)
return integer
def InterpretNumber(token):
number = InterpretInteger(token)
if number == None:
return InterpretHexInteger(token)
else:
return number
def InterpretBytes(token):
if token[0] == STATE_STRING:
return token[1]
if token[0] != STATE_IDENTIFIER:
return None
if not token[1].startswith('0x'):
return None
return Hex2Bytes(token[1][2:])
def CheckFunction(functionname, arguments, countarguments, maxcountarguments=None):
if maxcountarguments == None:
if countarguments == 0 and len(arguments) != 0:
print('Error: function %s takes no arguments, %d are given' % (functionname, len(arguments)))
return True
if countarguments == 1 and len(arguments) != 1:
print('Error: function %s takes 1 argument, %d are given' % (functionname, len(arguments)))
return True
if countarguments != len(arguments):
print('Error: function %s takes %d arguments, %d are given' % (functionname, countarguments, len(arguments)))
return True
else:
if len(arguments) < countarguments or len(arguments) > maxcountarguments:
print('Error: function %s takes between %d and %d arguments, %d are given' % (functionname, countarguments, maxcountarguments, len(arguments)))
return True
return False
def CheckNumber(argument, minimum=None, maximum=None):
number = InterpretNumber(argument)
if number == None:
print('Error: argument should be a number: %s' % argument[1])
return None
if minimum != None and number < minimum:
print('Error: argument should be minimum %d: %d' % (minimum, number))
return None
if maximum != None and number > maximum:
print('Error: argument should be maximum %d: %d' % (maximum, number))
return None
return number
def Interpret(expression):
functioncalls = Parse(expression)
if functioncalls == None:
return None
decoded = ''
for functioncall in functioncalls:
functionname, arguments = functioncall
if functionname == FUNCTIONNAME_REPEAT:
if CheckFunction(functionname, arguments, 2):
return None
number = CheckNumber(arguments[0], minimum=1)
if number == None:
return None
bytes = InterpretBytes(arguments[1])
if bytes == None:
print('Error: argument should be a byte sequence: %s' % arguments[1][1])
return None
decoded += number * bytes
elif functionname == FUNCTIONNAME_RANDOM:
if CheckFunction(functionname, arguments, 1):
return None
number = CheckNumber(arguments[0], minimum=1)
if number == None:
return None
decoded += ''.join([chr(random.randint(0, 255)) for x in range(number)])
elif functionname == FUNCTIONNAME_LOREMIPSUM:
if CheckFunction(functionname, arguments, 1):
return None
number = CheckNumber(arguments[0], minimum=1)
if number == None:
return None
decoded += LoremIpsum(number)
elif functionname == FUNCTIONNAME_CHR:
if CheckFunction(functionname, arguments, 1, 2):
return None
number = CheckNumber(arguments[0], minimum=0, maximum=255)
if number == None:
return None
if len(arguments) == 1:
decoded += chr(number)
else:
number2 = CheckNumber(arguments[1], minimum=0, maximum=255)
if number2 == None:
return None
if number < number2:
decoded += ''.join([chr(n) for n in range(number, number2 + 1)])
else:
decoded += ''.join([chr(n) for n in range(number, number2 - 1, -1)])
else:
print('Error: unknown function: %s' % functionname)
return None
return decoded
def ParsePackExpression(data):
try:
packFormat, pythonExpression = data.split('#', 1)
data = struct.pack(packFormat, int(pythonExpression))
return data
except:
return None
FCH_FILENAME = 0
FCH_DATA = 1
FCH_ERROR = 2
def FilenameCheckHash(filename, literalfilename):
if literalfilename:
return FCH_FILENAME, filename
elif filename.startswith('#h#'):
result = Hex2Bytes(filename[3:].replace(' ', ''))
if result == None:
return FCH_ERROR, 'hexadecimal'
else:
return FCH_DATA, result
elif filename.startswith('#b#'):
try:
return FCH_DATA, binascii.a2b_base64(filename[3:])
except:
return FCH_ERROR, 'base64'
elif filename.startswith('#e#'):
result = Interpret(filename[3:])
if result == None:
return FCH_ERROR, 'expression'
else:
return FCH_DATA, C2BIP3(result)
elif filename.startswith('#p#'):
result = ParsePackExpression(filename[3:])
if result == None:
return FCH_ERROR, 'pack'
else:
return FCH_DATA, result
elif filename.startswith('#'):
return FCH_DATA, C2BIP3(filename[1:])
else:
return FCH_FILENAME, filename
def AnalyzeFileError(filename):
PrintError('Error opening file %s' % filename)
PrintError(sys.exc_info()[1])
try:
if not os.path.exists(filename):
PrintError('The file does not exist')
elif os.path.isdir(filename):
PrintError('The file is a directory')
elif not os.path.isfile(filename):
PrintError('The file is not a regular file')
except:
pass
class cBinaryFile:
def __init__(self, filename, zippassword='infected', noextraction=False, literalfilename=False):
self.filename = filename
self.zippassword = zippassword
self.noextraction = noextraction
self.literalfilename = literalfilename
self.oZipfile = None
self.extracted = False
self.fIn = None
fch, data = FilenameCheckHash(self.filename, self.literalfilename)
if fch == FCH_ERROR:
line = 'Error %s parsing filename: %s' % (data, self.filename)
raise Exception(line)
try:
if self.filename == '':
if sys.platform == 'win32':
import msvcrt
msvcrt.setmode(sys.stdin.fileno(), os.O_BINARY)
self.fIn = sys.stdin
elif fch == FCH_DATA:
self.fIn = DataIO(data)
elif not self.noextraction and self.filename.lower().endswith('.zip'):
self.oZipfile = zipfile.ZipFile(self.filename, 'r')
if len(self.oZipfile.infolist()) == 1:
self.fIn = self.oZipfile.open(self.oZipfile.infolist()[0], 'r', self.zippassword)
self.extracted = True
else:
self.oZipfile.close()
self.oZipfile = None
self.fIn = open(self.filename, 'rb')
elif not self.noextraction and self.filename.lower().endswith('.gz'):
self.fIn = gzip.GzipFile(self.filename, 'rb')
self.extracted = True
else:
self.fIn = open(self.filename, 'rb')
except:
AnalyzeFileError(self.filename)
raise
def close(self):
if self.fIn != sys.stdin and self.fIn != None:
self.fIn.close()
if self.oZipfile != None:
self.oZipfile.close()
def read(self, size=None):
try:
fRead = self.fIn.buffer
except:
fRead = self.fIn
if size == None:
return fRead.read()
else:
return fRead.read(size)
def Data(self):
data = self.read()
self.close()
return data
#-ENDCODE cBinaryFile--------------------------------------------------------------------------------
def File2Strings(filename):
try:
if filename == '':
f = sys.stdin
else:
f = open(filename, 'r')
except:
return None
try:
return map(lambda line:line.rstrip('\n'), f.readlines())
except:
return None
finally:
if f != sys.stdin:
f.close()
def File2String(filename):
try:
f = open(filename, 'rb')
except:
return None
try:
return f.read()
except:
return None
finally:
f.close()
def ProcessAt(argument):
if argument.startswith('@'):
strings = File2Strings(argument[1:])
if strings == None:
raise Exception('Error reading %s' % argument)
else:
return strings
else:
return [argument]
def Glob(filename):
filenames = glob.glob(filename)
if len(filenames) == 0:
return [filename]
else:
return filenames
class cExpandFilenameArguments():
def __init__(self, filenames, literalfilenames=False, recursedir=False, checkfilenames=False, expressionprefix=None, flagprefix=None):
self.containsUnixShellStyleWildcards = False
self.warning = False
self.message = ''
self.filenameexpressionsflags = []
self.expressionprefix = expressionprefix
self.flagprefix = flagprefix
self.literalfilenames = literalfilenames
expression = ''
flag = ''
if len(filenames) == 0:
self.filenameexpressionsflags = [['', '', '']]
elif literalfilenames:
self.filenameexpressionsflags = [[filename, '', ''] for filename in filenames]
elif recursedir:
for dirwildcard in filenames:
if expressionprefix != None and dirwildcard.startswith(expressionprefix):
expression = dirwildcard[len(expressionprefix):]
elif flagprefix != None and dirwildcard.startswith(flagprefix):
flag = dirwildcard[len(flagprefix):]
else:
if dirwildcard.startswith('@'):
for filename in ProcessAt(dirwildcard):
self.filenameexpressionsflags.append([filename, expression, flag])
elif os.path.isfile(dirwildcard):
self.filenameexpressionsflags.append([dirwildcard, expression, flag])
else:
if os.path.isdir(dirwildcard):
dirname = dirwildcard
basename = '*'
else:
dirname, basename = os.path.split(dirwildcard)
if dirname == '':
dirname = '.'
for path, dirs, files in os.walk(dirname):
for filename in fnmatch.filter(files, basename):
self.filenameexpressionsflags.append([os.path.join(path, filename), expression, flag])
else:
for filename in list(collections.OrderedDict.fromkeys(sum(map(self.Glob, sum(map(ProcessAt, filenames), [])), []))):
if expressionprefix != None and filename.startswith(expressionprefix):
expression = filename[len(expressionprefix):]
elif flagprefix != None and filename.startswith(flagprefix):
flag = filename[len(flagprefix):]
else:
self.filenameexpressionsflags.append([filename, expression, flag])
self.warning = self.containsUnixShellStyleWildcards and len(self.filenameexpressionsflags) == 0
if self.warning:
self.message = "Your filename argument(s) contain Unix shell-style wildcards, but no files were matched.\nCheck your wildcard patterns or use option literalfilenames if you don't want wildcard pattern matching."
return
if self.filenameexpressionsflags == [] and (expression != '' or flag != ''):
self.filenameexpressionsflags = [['', expression, flag]]
if checkfilenames:
self.CheckIfFilesAreValid()
def Glob(self, filename):
if not ('?' in filename or '*' in filename or ('[' in filename and ']' in filename)):
return [filename]
self.containsUnixShellStyleWildcards = True
return glob.glob(filename)
def CheckIfFilesAreValid(self):
valid = []
doesnotexist = []
isnotafile = []
for filename, expression, flag in self.filenameexpressionsflags:
hashfile = False
try:
hashfile = FilenameCheckHash(filename, self.literalfilenames)[0] == FCH_DATA
except:
pass
if filename == '' or hashfile:
valid.append([filename, expression, flag])
elif not os.path.exists(filename):
doesnotexist.append(filename)
elif not os.path.isfile(filename):
isnotafile.append(filename)
else:
valid.append([filename, expression, flag])
self.filenameexpressionsflags = valid
if len(doesnotexist) > 0:
self.warning = True
self.message += 'The following files do not exist and will be skipped: ' + ' '.join(doesnotexist) + '\n'
if len(isnotafile) > 0:
self.warning = True
self.message += 'The following files are not regular files and will be skipped: ' + ' '.join(isnotafile) + '\n'
def Filenames(self):
if self.expressionprefix == None:
return [filename for filename, expression, flag in self.filenameexpressionsflags]
else:
return self.filenameexpressionsflags
def CheckJSON(stringJSON):
try:
object = json.loads(stringJSON)
except:
print('Error parsing JSON')
print(sys.exc_info()[1])
return None
if not isinstance(object, dict):
print('Error JSON is not a dictionary')
return None
if not 'version' in object:
print('Error JSON dictionary has no version')
return None
if object['version'] != 2:
print('Error JSON dictionary has wrong version')
return None
if not 'id' in object:
print('Error JSON dictionary has no id')
return None
if object['id'] != 'didierstevens.com':
print('Error JSON dictionary has wrong id')
return None
if not 'type' in object:
print('Error JSON dictionary has no type')
return None
if object['type'] != 'content':
print('Error JSON dictionary has wrong type')
return None
if not 'fields' in object:
print('Error JSON dictionary has no fields')
return None
if not 'name' in object['fields']:
print('Error JSON dictionary has no name field')
return None
if not 'content' in object['fields']:
print('Error JSON dictionary has no content field')
return None
if not 'items' in object:
print('Error JSON dictionary has no items')
return None
for item in object['items']:
item['content'] = binascii.a2b_base64(item['content'])
return object['items']
CUTTERM_NOTHING = 0
CUTTERM_POSITION = 1
CUTTERM_FIND = 2
CUTTERM_LENGTH = 3
def Replace(string, dReplacements):
if string in dReplacements:
return dReplacements[string]
else:
return string
def ParseInteger(argument):
sign = 1
if argument.startswith('+'):
argument = argument[1:]
elif argument.startswith('-'):
argument = argument[1:]
sign = -1
if argument.startswith('0x'):
return sign * int(argument[2:], 16)
else:
return sign * int(argument)
def ParseCutTerm(argument):
if argument == '':
return CUTTERM_NOTHING, None, ''
oMatch = re.match(r'\-?0x([0-9a-f]+)', argument, re.I)
if oMatch == None:
oMatch = re.match(r'\-?(\d+)', argument)
else:
value = int(oMatch.group(1), 16)
if argument.startswith('-'):
value = -value
return CUTTERM_POSITION, value, argument[len(oMatch.group(0)):]
if oMatch == None:
oMatch = re.match(r'\[([0-9a-f]+)\](\d+)?([+-](?:0x[0-9a-f]+|\d+))?', argument, re.I)
else:
value = int(oMatch.group(1))
if argument.startswith('-'):
value = -value
return CUTTERM_POSITION, value, argument[len(oMatch.group(0)):]
if oMatch == None:
oMatch = re.match(r"\[u?\'(.+?)\'\](\d+)?([+-](?:0x[0-9a-f]+|\d+))?", argument)
else:
if len(oMatch.group(1)) % 2 == 1:
raise Exception("Uneven length hexadecimal string")
else:
return CUTTERM_FIND, (binascii.a2b_hex(oMatch.group(1)), int(Replace(oMatch.group(2), {None: '1'})), ParseInteger(Replace(oMatch.group(3), {None: '0'}))), argument[len(oMatch.group(0)):]
if oMatch == None:
return None, None, argument
else:
if argument.startswith("[u'"):
# convert ascii to unicode 16 byte sequence
searchtext = oMatch.group(1).decode('unicode_escape').encode('utf16')[2:]
else:
searchtext = oMatch.group(1)
return CUTTERM_FIND, (searchtext, int(Replace(oMatch.group(2), {None: '1'})), ParseInteger(Replace(oMatch.group(3), {None: '0'}))), argument[len(oMatch.group(0)):]
def ParseCutArgument(argument):
type, value, remainder = ParseCutTerm(argument.strip())
if type == CUTTERM_NOTHING:
return CUTTERM_NOTHING, None, CUTTERM_NOTHING, None
elif type == None:
if remainder.startswith(':'):
typeLeft = CUTTERM_NOTHING
valueLeft = None
remainder = remainder[1:]
else:
return None, None, None, None
else:
typeLeft = type
valueLeft = value
if typeLeft == CUTTERM_POSITION and valueLeft < 0:
return None, None, None, None
if typeLeft == CUTTERM_FIND and valueLeft[1] == 0:
return None, None, None, None
if remainder.startswith(':'):
remainder = remainder[1:]
else:
return None, None, None, None
type, value, remainder = ParseCutTerm(remainder)
if type == CUTTERM_POSITION and remainder == 'l':
return typeLeft, valueLeft, CUTTERM_LENGTH, value
elif type == None or remainder != '':
return None, None, None, None
elif type == CUTTERM_FIND and value[1] == 0:
return None, None, None, None
else:
return typeLeft, valueLeft, type, value
def Find(data, value, nth, startposition=-1):
position = startposition
while nth > 0:
position = data.find(value, position + 1)
if position == -1:
return -1
nth -= 1
return position
def CutData(stream, cutArgument):
if cutArgument == '':
return [stream, None, None]
typeLeft, valueLeft, typeRight, valueRight = ParseCutArgument(cutArgument)
if typeLeft == None:
return [stream, None, None]
if typeLeft == CUTTERM_NOTHING:
positionBegin = 0
elif typeLeft == CUTTERM_POSITION:
positionBegin = valueLeft
elif typeLeft == CUTTERM_FIND:
positionBegin = Find(stream, valueLeft[0], valueLeft[1])
if positionBegin == -1:
return ['', None, None]
positionBegin += valueLeft[2]
else:
raise Exception("Unknown value typeLeft")
if typeRight == CUTTERM_NOTHING:
positionEnd = len(stream)
elif typeRight == CUTTERM_POSITION and valueRight < 0:
positionEnd = len(stream) + valueRight
elif typeRight == CUTTERM_POSITION:
positionEnd = valueRight + 1
elif typeRight == CUTTERM_LENGTH:
positionEnd = positionBegin + valueRight
elif typeRight == CUTTERM_FIND:
positionEnd = Find(stream, valueRight[0], valueRight[1], positionBegin)
if positionEnd == -1:
return ['', None, None]
else:
positionEnd += len(valueRight[0])
positionEnd += valueRight[2]
else:
raise Exception("Unknown value typeRight")
return [stream[positionBegin:positionEnd], positionBegin, positionEnd]
#-BEGINCODE cDump------------------------------------------------------------------------------------
#import binascii
#import sys
#if sys.version_info[0] >= 3:
# from io import StringIO
#else:
# from cStringIO import StringIO
class cDump():
def __init__(self, data, prefix='', offset=0, dumplinelength=16):
self.data = data
self.prefix = prefix
self.offset = offset
self.dumplinelength = dumplinelength
def HexDump(self):
oDumpStream = self.cDumpStream(self.prefix)
hexDump = ''
for i, b in enumerate(self.data):
if i % self.dumplinelength == 0 and hexDump != '':
oDumpStream.Addline(hexDump)
hexDump = ''
hexDump += IFF(hexDump == '', '', ' ') + '%02X' % self.C2IIP2(b)
oDumpStream.Addline(hexDump)
return oDumpStream.Content()
def CombineHexAscii(self, hexDump, asciiDump):
if hexDump == '':
return ''
countSpaces = 3 * (self.dumplinelength - len(asciiDump))
if len(asciiDump) <= self.dumplinelength / 2:
countSpaces += 1
return hexDump + ' ' + (' ' * countSpaces) + asciiDump
def HexAsciiDump(self, rle=False):
oDumpStream = self.cDumpStream(self.prefix)
position = ''
hexDump = ''
asciiDump = ''
previousLine = None
countRLE = 0
for i, b in enumerate(self.data):
b = self.C2IIP2(b)
if i % self.dumplinelength == 0:
if hexDump != '':
line = self.CombineHexAscii(hexDump, asciiDump)
if not rle or line != previousLine:
if countRLE > 0:
oDumpStream.Addline('* %d 0x%02x' % (countRLE, countRLE * self.dumplinelength))
oDumpStream.Addline(position + line)
countRLE = 0
else:
countRLE += 1
previousLine = line
position = '%08X:' % (i + self.offset)
hexDump = ''
asciiDump = ''
if i % self.dumplinelength == self.dumplinelength / 2:
hexDump += ' '
hexDump += ' %02X' % b
asciiDump += IFF(b >= 32 and b < 127, chr(b), '.')
if countRLE > 0:
oDumpStream.Addline('* %d 0x%02x' % (countRLE, countRLE * self.dumplinelength))
oDumpStream.Addline(self.CombineHexAscii(position + hexDump, asciiDump))
return oDumpStream.Content()
def Base64Dump(self, nowhitespace=False):
encoded = binascii.b2a_base64(self.data).decode().strip()
if nowhitespace:
return encoded
oDumpStream = self.cDumpStream(self.prefix)
length = 64
for i in range(0, len(encoded), length):
oDumpStream.Addline(encoded[0+i:length+i])
return oDumpStream.Content()
class cDumpStream():
def __init__(self, prefix=''):
self.oStringIO = StringIO()
self.prefix = prefix
def Addline(self, line):
if line != '':