LATIDIA · Ciberseguridad

CASHEWS: Preprocesador de origen para la detección de paquetes maliciosos basados en LLM

arXiv:2609.18862v1 Announce Type: new Resumen: Las herramientas de detección de paquetes npm maliciosos ahora aprovechan la comprensión semántica del código fuente de los LLM para detectar intenciones maliciosas a escala. Esta capacidad ha demostrado ser invaluable en

La noticia

arXiv:2609.18862v1 Announce Type: new Abstract: Malicious npm package detection tools now leverage LLMs' semantic understanding of source code to detect malicious intent at scale. This capability has proven invaluable in identifying packages involved in recent supply-chain attacks such as Shai-Hulud. However, threat actors exploit the limited context windows of LLMs through JavaScript techniques such as code obfuscation that yields high token density and bundling malicious code with benign packages, causing detectors to skip large files or miss malicious behavior. This creates an attack surface for evading detection. In this paper, we present CASHEWS, a JavaScript preprocessor that reduces file size by rewriting source code to remove code that is irrelevant