LATIDIA · Ciberseguridad
ToolPulse: espionaje por canal lateral de agentes LLM que utilizan herramientas
arXiv:2611.25470v1 Tipo de anuncio: nuevo Resumen: Los grandes modelos de lenguaje (LLM) se implementan cada vez más como agentes que traducen las solicitudes de los usuarios en secuencias de invocaciones de herramientas locales. Si bien los ataques por canal lateral basados en la caché previos
WhatsApp ↗Telegram ↗
La noticia
arXiv:2611.25470v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed as agents that translate user requests into sequences of local tool invocations. While prior cache-based side-channel attacks target the locally deployed models, they typically require the adversary to share model-related components or hardware resources with the victim's inference process. These assumptions do not apply to the predominant agentic settings, in which commercial LLMs execute on vendor-controlled infrastructure. We show that agentic execution exposes a more practical attack surface: although inference remains remote, tools are executed on the user's machine and commonly rely on shared system libraries. In this paper, we first systematically evaluate Flush+Reload and Prime+Probe for monitoring