# 安全 · 日跟踪 · 2026-09-29 > 日期:2026-09-29(Asia/Shanghai)。专题正文:`intel/topics/security.md`。原料:`/workspace/intel/archive/2026-09-29/`;对照 2026-09-28 日跟踪。 ## 摘要 - **NEW · RH securing AI agents + OpenShell onboarding(系统周边):** (1)**Securing AI agents requires securing the systems around them**——身份/工具/运行时/网络/OS 纵深;OpenShift AI / OpenShift / RHEL 分层;OpenShell + BlueField 为独立执行点;(2)**Why Red Hat is building secure agent onboarding**——OpenShell 开源安全运行时(Landlock/seccomp/netns/L7;凭证边界注入)——**厂商博客 NEW;OpenShell 仍开源/DP 叙事,勿写成 ACS/OpenShell GA**。 - **续档 ·** RHEL 10 STIG V1R2;AAP security risk;Blueprint / Open Secure AI→LF / Lightwell / Rubrik×CrowdStrike / OCP ML-KEM / VCF hardening。 - **不抬升:** 将 OpenShell 写成已 GA 的 ACS 对标产品,或把 STIG 再抬成 OCP 发行说明。 ## 原文摘录 ### RH · Securing AI agents(NEW) - 归档:`redhat/securing-ai-agents-requires-securing-the-systems-around-them-79bd2c93.*` - URL:https://www.redhat.com/en/blog/securing-ai-agents-requires-securing-systems-around-them > … securing the **system** in which the agent acts… > … **OpenShift AI**… **OpenShift**… **RHEL**… > … **OpenShell**… **NVIDIA BlueField** DPUs… **Lightwell**… Open Secure AI Alliance… ### RH · Secure agent onboarding / OpenShell(NEW) - 归档:`redhat/why-red-hat-is-building-secure-agent-onboarding-e125232b.*` - URL:https://www.redhat.com/en/blog/why-red-hat-is-building-secure-agent-onboarding > … **OpenShell** — an **open source** project and a secure agent runtime… NVIDIA **Open Agent Safety Platform**… > … **Landlock**, **seccomp**… network **namespace**… **L7**… Credentials … outside … injected … at the network boundary… ### 续档指针 - RHEL STIG / AAP:见 `tracking/topics/security/2026-09-28.md`。 ### 检索缺口 - `docs.redhat.com` 仍常 blocked;无 ACS 发行说明新稿;OpenShell 无独立 GA 公告。