Here's how a protection test works. The microcode feeds three things to the Test PLA:
第一百二十条 当场作出治安管理处罚决定的，人民警察应当向违反治安管理行为人出示人民警察证，并填写处罚决定书。处罚决定书应当当场交付被处罚人；有被侵害人的，并应当将决定书送达被侵害人。
，更多细节参见WPS下载最新地址
Продажи коммерческих автомобилей в России рухнули«Автостат»: Продажи коммерческих автомобилей в РФ в 2025 году упали на 37,5 %
Driver size: 11mm
The common pattern across all of these seems to be filesystem and network ACLs enforced by the OS, not a separate kernel or hardware boundary. A determined attacker who already has code execution on your machine could potentially bypass Seatbelt or Landlock restrictions through privilege escalation. But that is not the threat model. The threat is an AI agent that is mostly helpful but occasionally careless or confused, and you want guardrails that catch the common failure modes - reading credentials it should not see, making network calls it should not make, writing to paths outside the project.