Awesome
QuickSand Version 2
QuickSand Python Package and Command Line Tool
QuickSand is a Python-based analysis framework to analyze suspected malware documents to identify exploits in streams of different encodings or compressions. QuickSand supports documents, PDFs, Mime/Email, Postscript and other common formats. A built-in command line tool can process a single document or directory of documents.
QuickSand scans within the decoded streams of documents and PDFs using Yara signatures to identify exploits or high risk active content.
A hosted version is available to try without any installation at scan.tylabs.com.
Files:
-
src/quicksand/quicksand.py: Main quicksand class and CLI tool
-
src/quicksand/quicksand_exe.yara: Yara rules to detect executables.
-
src/quicksand/quicksand_exploits.yara: Yara rules to detect exploits in documents.
-
src/quicksand/quicksand_pdf.yara: Yara rules to detect exploits in PDFs.
-
bin/quicksand: Command line tool.
-
requirements.txt: Python dependencies
-
lambda/ Optional AWS Lambda functions
With Thanks to the Creators of:
-
pdfreader
-
oletools
-
cryptography
-
zipfile38
-
olefile
-
yara-python
-
yara
Installation from Pypi using pip
pip3 install quicksand
Upgrade from Pypi using pip
pip3 install --upgrade quicksand
Install from source
If you want to install from the source, such as the uicksand-main.zip downloaded from GitHub:
pip3 install quicksand-main.zip
Command Line Usage
A command line tool for quicksand to process and output json or txt results.
usage: quicksand [-h] [-v] [-c] [-y] [-t TIMEOUT] [-e EXPLOIT] [-x EXE] [-a PDF] [-f {json,txt}] [-o OUT] [-p PASSWORD]
[-d DROPDIR]
document
QuickSand Document and PDF maldoc analysis tool.
positional arguments:
document document or directory to scan
optional arguments:
-h, --help show this help message and exit
-v, --verbose increase output verbosity
-c, --capture capture stream content
-y, --yara capture yara matched strings
-t TIMEOUT, --timeout TIMEOUT
timeout in seconds
-e EXPLOIT, --exploit EXPLOIT
yara exploit signatures
-x EXE, --exe EXE yara executable signatures
-a PDF, --pdf PDF yara PDF signatures
-f {json,txt}, --format {json,txt}
output format
-o OUT, --out OUT save output to this filename
-p PASSWORD, --password PASSWORD
password to decrypt ole or pdf
-d DROPDIR, --dropdir DROPDIR
save objects to this directory
Process a single file
quicksand document.doc
Process a directory of files
quicksand malware/
Python Module Usage
File from memory
from quicksand.quicksand import quicksand
import pprint
qs = quicksand(data, timeout=18, strings=True)
qs.process()
pprint.pprint(qs.results)
Processing using a filename
from quicksand.quicksand import quicksand
qs2 = quicksand("file.doc")
qs2.process()
qs.results
Process a Directory
from quicksand.quicksand import quicksand
qs = quicksand.readDir("malware")
qs
Returns a dictionary of {filename: qs_results
,...}.
Optional initializer values
-
capture: True|False return content of extracted streams
-
debug: True|False print debugging messages to stdout
-
exploityara: Path to exploit yara rules
-
execyara: Path to executable yara rules
-
pdfyara: PDF Exploits yara rules
-
password: Password for encrypted documents/PDFs
-
timeout: Timeout processing: 0 for unlimited.
Scoring
Documents are scored based on the rank value in the associated Yara signature metadata.
Additionally, each signature defines whether the detected item is an exploit, a warning or a risky feature. For more information on how to interpret the results, please see https://scan.tylabs.com/howto.
If you add your own signatures, they don't need to include the extra metadata to function.
zlib issues on MacOS
MacOS users may get zlib issues (PDF FlateDecode etc) due to missing OpenSSL headers since MacOs 10.4.
zlib.error: Error -3 while decompressing data: unknown compression method
zlib.error: Error -3 while decompressing data: incorrect header check
One solution is to install zlib with Brew.sh and reinstall Python 3 using pyenv:
export LDFLAGS="-L/usr/local/opt/zlib/lib"
export CPPFLAGS="-I/usr/local/opt/zlib/include"
pyenv install 3.8.5
Using Quicksand?
Let us know @tylabs