LATIDIA · Ciberseguridad

The Devil Wears Markdown: Detecting Malicious Intent Hidden in AI Agent Skill Instructions through Hybrid Static, Dynamic, and Agentic Analysis

arXiv: 2611.10830v1Tipo de anuncio: nuevo Resumen: Los ecosistemas genéticos de IA dependen cada vez más de habilidades de agentes reutilizables que combinan código ejecutable, instrucciones en lenguaje natural, artefactos de configuración e int. de herramientas externas

La noticia

arXiv:2611.10830v1 Announce Type: new Abstract: Agentic AI ecosystems increasingly rely on reusable agent skills that combine executable code, natural-language instructions, configuration artifacts, and external tool integrations. These hybrid artifacts create software supply-chain risks because malicious intent may be distributed across instructions, code, metadata, and runtime behavior, making isolated alerts difficult to interpret. We evaluate three open-source scanners on a held-out test set of 140 skills comprising 80 malicious and 60 benign samples, including Cisco Skill Scanner with and without its optional LLM analyzer. The default Cisco configuration achieves a recall of 0.91 with 33 false positives among the 60 benign skills, while the LLM-enabled configuration using Qwen3:8B achieves a recall