Skip to content

Latest commit

 

History

History
18 lines (13 loc) · 1.73 KB

WHY.md

File metadata and controls

18 lines (13 loc) · 1.73 KB

Why to make another detector if there is already one?


Unlike the Neko Detector, I wanted to make something more future-proof.

Comparison:

jNeedle Neko Detector
Target It is a generalized tool designed to detect any JVM malware Specific for Skyrage and fractureiser
Reliability Strict signature sequence matching in function, it should produce less false-positive Custom scanning for each rule, more reliable in detecting fractureiser specifically.
Speed Uses an efficient KMP sequence matching algorithm with custom rules, O(n+m) Custom scan algorithm, slower (~2.5x slower than jNeedle with only a few rules)
New virus detection I could add a new signature in less than 25 minutes after receiving the sample New custom rule has to be added, way slower (maybe hours)
Database update Online database is updated automatically, updating jNeedle is not required. New program release has to be made and update is required for new detections.

jNeedle is a much better approach than Neko Detector, but was slower to implement. It's an improvement over the existing tool.