This is the CrowdStrike backend for pySigma. It provides the package sigma.backends.crowdstrike with the LogScaleBackend class.

Further it contains the following processing pipelines:

• crowdstrike_fdr_pipeline which was mainly written for the Falcon data Replicator data but Splunk queries should work in the legacy CrowdStrike Splunk
• crowdstrike_falcon_pipeline which was written for data collected by the CrowdStrike Falcon Agent stored in CrowdStrike Logscale. It effectively translates rules to the CrowdStrike Query Language used by LogScale.

The following categories and products are supported by the crowdstrike_falcon_pipeline pipeline:

The following categories and products are supported by the crowdstrike_fdr_pipeline pipeline:

There's likely more windows categories that can be supported by the pipelines; We will be adding support gradually as availability allows.

• Full Paths: Falcon agents do not capture drive names when logging paths. Instead, when drive letters are expected the device path is used. For example, C:\Windows results to \Device\HarddiskVolume3\Windows in the logs. To account for this, the pipeline replaces any drive letters in fields containing full path with \Device\HarddiskVolume?\ (where '?' can be any single character).

• Parent Name: Falcon process_creation events do not capture the full path of the parent. Hence, in such cases the transformation is configured to fail.

• DNS Query Results: Falcon dns_query events return the IP records of a successful query in semicolon-separated string. The pipeline handles this by enforcing a "contains" expression on the QueryResults field

• Unsupported fields: Falcon does not always capture the same fields as sysmon for the categories supported. In cases where the rule requires unsupported fields, the transformation fails.

• LogScale Community Content

This backend is currently maintained by:

• Thomas Patzke
• Panos Moullotos