LATIDIA · Ciberseguridad

AgentLSD: Evaluación de agentes de seguridad de IA bajo contaminación de tareas adversas

arXiv:2609.19140v1 Tipo de anuncio: nuevo Resumen: los agentes de IA para la seguridad inspeccionan páginas web, código fuente, registros, archivos de configuración y salidas de comandos. Estos entornos pueden contener artefactos engañosos que influyen en la

La noticia

arXiv:2609.19140v1 Announce Type: new Abstract: AI agents for security inspect web pages, source code, logs, configuration files, and command outputs. These environments may contain deceptive artifacts that influence the agent's behavior. We call this adversarial task contamination. Whereas prompt injection relies on attacker-supplied instructions, task contamination also includes non-instructional evidence, such as fake results and decoy endpoints. We present AgentLSD, a controlled framework for studying adversarial task contamination. AgentLSD uses Capture the Flag (CTF) challenges as its experimental environment. We inject trap artifacts, such as fake flags, misleading hints, decoy endpoints, and hidden cues, while preserving the intended CTF solution. The framework supports paired clean and trap-augmented experiments with deterministic