LATIDIA · Ciberseguridad

SecureClaw: Clawing Back Control of LLM Agents

arXiv: 2606.09549v2Tipo de anuncio: reemplazar Resumen: Los agentes del modelo de lenguaje grande (LLM) que utilizan herramientas se enfrentan a dos fallas de seguridad distintas: acciones externas no autorizadas y exposición de texto plano sensible dentro del tiempo de ejecución b

La noticia

arXiv:2606.09549v2 Announce Type: replace Abstract: Tool-using large language model (LLM) agents face two distinct security failures: unauthorized external actions and exposure of sensitive plaintext inside the runtime before any final output check can intervene. Existing defenses usually protect one boundary, either the planner/runtime or the point where external actions are executed, and therefore do not by themselves secure both surfaces. We present SecureClaw, a dual-boundary architecture that enforces authorization in a trusted executor and keeps raw protected data outside the LLM runtime. Sensitive reads pass through a trusted gateway that replaces raw values with reference IDs. In the evaluated deployment, the gateway also provides short planning summaries that deliberately release