Skip to content

decalage2/ViperMonkey

Repository files navigation

ViperMonkey

ViperMonkey is a VBA Emulation engine written in Python, designed to analyze and deobfuscate malicious VBA Macros contained in Microsoft Office files (Word, Excel, PowerPoint, Publisher, etc).

See the article "Using VBA Emulation to Analyze Obfuscated Macros", for real-life examples of malware deobfucation with ViperMonkey.

ViperMonkey was also demonstrated at the Black Hat Europe 2019 conference: see the slides and video (at 18:38).

ViperMonkey was created by Philippe Lagadec in 2015-2016, and the project is maintained in the repository https://github.com/decalage2/ViperMonkey. Since November 2017, most of the development is done by Kirk Sayre and other contributors in the repository https://github.com/kirk-sayre-work/ViperMonkey. The main repository is synchronised regularly, but cutting edge improvements are usually available first in Kirk's version.

Quick links: Report Issues/Suggestions/Questions - Contact the Author - Repository - Updates on Twitter - API Tutorial

DISCLAIMER:

  • ViperMonkey is an experimental VBA Engine targeted at analyzing maldocs. It works on some but not all maldocs.
  • VBA parsing and emulation is extremely slow for now (see the speedup section for how to improve the speed).
  • VBA Emulation is hard and complex, because of all the features of the VBA language, of Microsoft Office applications, and all the DLLs and ActiveX objects that can be called from VBA.
  • This open-source project is only developed on my scarce spare time, so do not expect miracles. Any help from you will be very appreciated!

Download and Install:

Easy Install

  1. Install docker.
  2. Run docker/dockermonkey.sh MYFILE to analyze file MYFILE.

dockermonkey.sh wil automatically pull down a preconfigured docker container, update ViperMonkey to the latest version in the container, and then analyze MYFILE by running ViperMonkey in the container. No other packages or configuration will need to be performed.

For information on using dockermonkey.sh run docker/dockermonkey.sh -h.

Installation using PyPy (recommended)

For performance reasons, it is highly recommended to use PyPy (5x faster), but it is also possible to run Vipermonkey with the normal Python interpreter (CPython) if you cannot use PyPy.

  1. If PyPy is not installed on your system, see http://pypy.org/download.html and download PyPy 2.7. (not 3.x)
  2. Check if pip is installed for pypy: run pypy -m pip
  3. If pip is not installed yet, run pypy -m ensurepip on Windows, or sudo -H pypy -m ensurepip on Linux/Mac
  4. Make sure pip is up-to-date, by running pypy -m pip install -U pip
  5. Download the archive from the repository: https://github.com/decalage2/ViperMonkey/archive/master.zip
  6. Extract it in the folder of your choice, and open a shell/cmd window in that folder.
  7. Under Ubuntu install pypy-dev (sudo apt-get install pypy-dev).
  8. Install dependencies by running pypy -m pip install -U -r requirements.txt on Windows, or sudo -H pypy -m pip install -U -r requirements.txt on Linux/Mac
  9. Check that Vipermonkey runs without error: pypy vmonkey.py

Installation using CPython

  1. Make sure you have the latest Python 2.7 installed: https://www.python.org/downloads/
  2. If you have both Python 2 and 3 versions installed, use pip2 instead of pip in the following commands, to install in Python 2 and not 3.
  3. Make sure pip is up-to-date, by running pip install -U pip
  4. Use pip to download and install vipermonkey with all its dependencies, by running the following command on Windows:
pip install -U https://github.com/decalage2/ViperMonkey/archive/master.zip

On Linux/Mac:

sudo -H pip install -U https://github.com/decalage2/ViperMonkey/archive/master.zip
  1. Check that Vipermonkey runs without error: open a shell/cmd window in any directory, an simply run vmonkey

Usage:

To run ViperMonkey in a Docker container with the -s, --jit, and --iocs options do:

docker/dockermonkey.sh <file>

To parse and interpret VBA macros from a document, use the vmonkey script:

vmonkey.py <file>

To make analysis faster (see the Speedup section), do:

pypy vmonkey.py -s <file>

Note: It is recommended to always use the -s option. When given the -s option ViperMonkey modifies some difficult to parse Visual Basic language constructs so that the ViperMonkey parser can correctly parse the input.

If the output is too verbose and too slow, you may reduce the logging level using the -l option:

vmonkey.py -l warning <file>

If the sample being analyzed has long running loops that are causing emulation to be unacceptably slow, use the --jit option to convert VB loops directly to Python in a JIT fashion during emulation.

vmonkey.py --jit <file>

Note: ViperMonkey's Python JIT loop conversion converts VB loops to Python and evals the generated Python code. While the Python conversion process is based on the parsed AST (not directly on the VB text) and VB data values are escaped/converted/modified to become valid in Python, any use of eval in Python potentially introduces a security risk. If this is a concern the dockermonkey.sh script can be used to run ViperMonkey in a sandboxed manner. dockermonkey.sh runs ViperMonkey in a fresh Docker container on each run (no file system modifications persist between runs) and networking is turned off in the Docker container.

Sometimes a malicious VBScript or Office file will generate IOCs during execution that are not used or that ViperMonkey does not see used. These intermediate IOCs are tracked by ViperMonkey during the emulation process and can be reported with the --iocs option.

vmonkey --iocs <file>

Note that one of the intermediate IOCs reported by ViperMonkey is injected shell code bytes. If the sample under analysis performs process injection directly in VB, ViperMonkey will report the injected byte values as an intermediate IOC with the --iocs flag. These byte values can then be written into a raw shell code file which can be further analyzed with a shell code emulator.

oletools Version

ViperMonkey requires the most recent version of oletools, at least v0.52.3. Make sure to either install the most recent oletools version by running pip install -U oletools, or make sure the most recent oletools install directory appears in PYTHONPATH, or install the most recent development version of oletools using pip as described here.

Speedup

pypy

The parsing library used by default in ViperMonkey can take a long time to parse some samples. ViperMonkey can be sped up considerably (~5 times faster) by running ViperMonkey using pypy rather than the regular Python interpreter. To use pypy do the following:

  1. Install pypy following the instructions here.
  2. Install the following Python packages. This can be done by downloading the .tar.gz for each package and running 'sudo pypy setup.py install' (note the use of pypy rather than python) for each package.
    1. setuptools
    2. colorlog
    3. olefile
    4. prettytable
    5. pyparsing

Stripping Useless Statements

The "-s" ViperMonkey command line option tells VipeMonkey to strip out useless statements from the Visual Basic macro code prior to parsing and emulation. For some maldocs this can significantly speed up analysis.

Emulating File Writes

ViperMonkey emulates some file writing behavior. The SHA256 hash of dropped files is reported in the ViperMonkey analysis results and the actual dropped files are saved in the directory MALDOC_artifacts/, where MALDOC is the name of the analyzed maldoc file.

ViperMonkey also searches Office 97 and Office 2007+ files for embedded PE files. These are automatically extracted and reported as dropped files in the MALDOC_artifacts/ directory.

Emulating Specific VBA Functions

By default ViperMonkey emulates maldoc behavior starting from standard macro auto run function (like AutoOpen, Document_Open, Document_Close, etc.). In some cases you may want to emulate the behavior starting from a non-standard auto run function. This is supported via the -i command line option. To emulate maldoc behavior starting from function Foo, use the command line option '-i Foo'. To emulate behavior starting from multiple non-standard entry points, use the command line option '-i "Foo,Bar,Baz"' (note that the entry point function names are comma seperated and must appear in a double quoted string).

API Interface:

ViperMonkey also includes a Python API interface that can be used for finer control emulation of your sample or for integration into an existing project.

Please see the API Tutorial for more information.

News

  • 2018-03-22 v0.06: new features and bug fixes contributed by Kirk Sayre
  • 2018-3:
    • Added support for parsing some technically invalid VBA statements.
    • Additional parsing fixes.
    • Added support for starting emulation at non-standard functions.
  • 2018-2:
    • Added support for Environ, IIf, Base64DecodeString, CLng, Close, Put, Run, InStrRev, LCase, RTrim, LTrim, AscW, AscB, and CurDir functions.
  • 2018-1
    • Added emulation support for saving dropped files.
    • Added support for For Each loops.
    • Added support for While Wend loops.
    • Handle 'Exit Do' instructions.
  • 2018-01-12 v0.05: a lot of new features and bug fixes contributed by Kirk Sayre
  • 2017-12-15:
    • Added support for Select and Do loops.
    • Added support for 'End Sub' and 0 argument return statements.
    • Added support for #if constructs.
    • Each VBA stream is now parsed in a separate thread (up to the # of machine cores).
  • 2017-11-28:
    • Added parsing for private type declarations.
    • Report calls to CreateProcessA in final report.
    • Handle Application.Run() of locally defined methods.
  • 2017-11-23:
    • Added VBA functions Abs, Fix, Hex, String, CByte, Atn, Dir, RGB, Log, Cos, Exp, Sin, Str, and Val.
    • Added support for 'Exit Function' operator.
    • Changed math operators to also work with string representations of integers.
    • Added a configurable iteration limit on loops.
  • 2017-11-14:
    • Added support for InStr, Replace, Sgn, Sqr, UBound, LBound, Trim, StrConv, Split, StrReverse, and Int VB functions.
    • Added support for string character subscripting.
    • Added support for negative integer literals.
    • Added support for if-then-else statements.
    • Added support for Const and initial values for global variable declarations.
    • Handle assignments of boolean expressions to variables.
  • 2017-11-03:
    • Added support for Left(), Right(), Array(), and BuiltInDocumentProperties() functions.
    • Added support for global variables.
    • Fixed some parse errors.
    • Added analysis of AutoClose() functions.
  • 2016-09-26 v0.02: First published version
  • 2015-02-28 v0.01: First development version
  • see changelog in source code for more info.

How to Suggest Improvements, Report Issues or Contribute:

This is a personal open-source project, developed on my spare time. Any contribution, suggestion, feedback or bug report is welcome.

To suggest improvements, report a bug or any issue, please use the issue reporting page, providing all the information and files to reproduce the problem.

You may also contact the author directly to provide feedback.

The code is available in a GitHub repository. You may use it to submit enhancements using forks and pull requests.

License

This license applies to the ViperMonkey package, apart from the thirdparty folder which contains third-party files published with their own license.

The ViperMonkey package is copyright (c) 2015-2020 Philippe Lagadec (http://www.decalage.info)

All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  • Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
  • Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.