Gisson/SS1617

A static analysis tool written in Python for identifying data flow integrity violations in PHP code.

SS1617

A static analysis tool written in Python for identifying data flow integrity violations in PHP code.

Dependencies

• phply (Python PHP parser)

  git clone --depth 1 https://github.com/viraptor/phply.git
  export PYTHONPATH="$PYTHONPATH:$(pwd)/phply"

  • PLY (Python Lex-Yacc) (needed by phply):

    pip install ply

    Make sure it's installed for the correct python version. To do run the command pip show ply and check the Location field.

Running the project

Make sure all the dependencies are properly installed. Then, try analysing one of the sample files:

cd src
./analyzer.py ../tests/sqli_02.php 2>/dev/null

2>/dev/null disables debug messages.

It should output something like:

> Tainted sink for SQL injection in line 3:
$hasil=mysql_query($q_sems,$koneksi);

Another example:

./analyzer.py ../tests/qli_02_sanitized.php 2>/dev/null

Should output something like:

> Sanitization function for SQL injection in line 2:
$nis=mysql_real_escape_string($nis);

Data structures

The data structure used is rule. This structure has attributes name (name of the rule), entry_point ( a possible entry point of a vulnerability), validation (the validation function used for the vulnerability) and lastly a sink (the sink of the problem, where the vulnerability can be executed). All these attributes but the name one are lists which are all the attributes encontered in a config file which is read by the analyzer. This attributes if more than 1 are divided by commas (,). See more about config files on the examples at the tests folder.

	list(struct:rule(string:name,list:entry_point,list:validation,list:sink))