LATIDIA · Ciberseguridad
Sopesar la envolvente: peso de los componentes y arbitraje en las defensas de capas de envolvente para un único agente LLM
arXiv:2611.23844v1 Tipo de anuncio: nuevo Resumen: La capa de envolvente es el canal a través del cual el contenido malicioso entra en un agente LLM. Trabajos anteriores establecieron los efectos de la etiqueta y la memoria en las envolturas A2A multiagente, pero el
WhatsApp ↗Telegram ↗
La noticia
arXiv:2611.23844v1 Announce Type: new Abstract: The envelope layer is the channel through which malicious content enters an LLM agent. Prior work established label and memory effects on multi-agent A2A envelopes, but the weight structure when MCP tools and memory coexist in a single agent remains unclear. Our threat model assumes an attacker pre-plants a malicious tool and poisons long-term memory; a benign user request inadvertently triggers it, and the attacker never enters the conversation. Following GDA Measurement and A/B/C/D, we measure how labels, memory, tool roles, and the backend shape invocation decisions across label-memory conflict, tool roles, robustness, and tool generalization. A tool-side forced-verification label substantially reduces semantic breach, but