Skip to content

About

Python tool for evaluating Sigma rules against events

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Latest commit

 

History

10 Commits

Folders and files

Repository files navigation

sigma-rule-matcher

sigma-rule-matcher is a Python package for evaluating Sigma detection rules against structured event data. Built on top of pySigma, it parses and applies Sigma rule logic—including condition expressions and most common modifiers—to incoming events to determine whether they match.

So far, this project is primarily a learning tool, put together to better understand how Sigma rules operate under the hood.

Installation

Install the package from PyPI:

pip install sigma-rule-matcher

Requires Python 3.10 or later. This will install the library and its dependencies:

  • boolean.py (for boolean expression evaluation)
  • pySigma (for parsing Sigma rules)

Usage

Sigma rules can include multiple selectors, logical operators, and modifiers. For example:

from sigma.rule import SigmaRule
from sigma_rule_matcher import RuleMatcher

sigma_rule = '''
title: Suspicious activity
logsource:
  product: test
detection:
  sel1:
    src_ip:
      - 10.0.0.1
      - 10.0.0.2
  sel2:
    user:
      - root
  sel3:
    process_name:
      - 'malicious.exe'
  condition: (sel1 or sel2) and sel3
'''

rule = SigmaRule.from_yaml(sigma_rule)
matcher = RuleMatcher(rule)

# Test against an event
event = {
    'src_ip': '10.0.0.2',
    'user': 'guest',
    'process_name': 'malicious.exe'
}

assert matcher.match(event) is True

Running Tests

The library includes a comprehensive test suite. To run the tests:

pytest

License

This project is licensed under the MIT License.

It uses the pySigma library, which is licensed under the GNU Lesser General Public License v2.1 (LGPL-2.1). A copy of the LGPL-2.1 license is here.

We use pySigma without modification.

About

Python tool for evaluating Sigma rules against events

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages