Agent libOS：能力受控的自我进化智能体运行时 Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents

精读摘要

LLM 智能体如今可以跨任务持久化、积累记忆、激活技能、合成工具、派生子进程、挂载远程资源、把检查点提交为可复用镜像——这些机制在部署后扩大了行动面，也带来权限升级与数据外泄的风险，因为「可见」常被误当成「有权限」。Agent libOS 是一个智能体原生的库操作系统基座，把操作准入、任务权限上限与类型化能力三个平面分离，从运行时层面约束智能体的自我进化。 LLM agents can persist, acquire memory, activate skills, synthesize tools, fork processes, and commit checkpoints, expanding the post-deployment action surface and creating authority-escalation and data-exfiltration risks when visibility is mistaken for permission. Agent libOS is an agent-native library OS substrate that separates operation admission, Task Authority ceilings, and typed Capabilities into distinct planes. It constrains self-evolving agents at the runtime level.