Host-based behavioral detection engine for Windows LOLBins, correlating Sysmon process telemetry with sequential attack chains, masquerading checks, and SOC alerting.
An enterprise-grade, host-based behavioral detection engine built in Python 3.12 for identifying Living-off-the-Land Binaries, Scripts, and Libraries (LOLBins) on Microsoft Windows endpoints.
Modern adversaries and post-exploitation frameworks increasingly evade traditional signature-based antivirus and Endpoint Detection and Response (EDR) solutions by abusing legitimate, digitally signed operating system utilities. This adversary tradecraft—commonly termed Living-off-the-Land (LotL)—allows attackers to execute code, download stages, bypass application whitelisting, and perform lateral movement without dropping custom binaries on disk.
lolbins-detector is a modular, standalone behavioral detection framework designed to detect anomalous utility invocations, renamed binaries, and multi-stage attack sequences in real time. It pairs high-fidelity process telemetry collection (preferring Sysmon Event ID 1 with automatic psutil fallback) with an in-memory correlation state machine, temporal event reconstruction (± 5–10 minutes), false positive suppression, an interactive SOC web dashboard, and automated incident reporting.
Relying solely on static indicators of compromise (such as binary file hashes or standard image paths) is fundamentally ineffective against LOLBins:
certutil.exe, mshta.exe, regsvr32.exe, rundll32.exe) are Microsoft-signed system binaries located in standard Windows directories (System32, SysWOW64).update.exe or svchost.exe). Checking file names alone is bypassed trivially.lolbins-detector counters these tactics by focusing on behavioral characteristics, PE OriginalFileName metadata, process tree lineage, and temporal sequence correlation.
The engine employs a decoupled, pipeline-driven architecture separating telemetry collection, correlation state tracking, rule evaluation, and artifact dispatch.