https://daoxuanshan.online/tags/%E9%98%B2%E6%8A%A4/ Recent content in 防护 on 忘川 Hugo -- gohugo.io zh-cn © 2026 忘川 Tue, 10 Mar 2026 00:00:00 +0000 https://daoxuanshan.online/docs/openclaw-skill-08-prompt-guard%E6%8F%90%E7%A4%BA%E6%B3%A8%E5%85%A5%E9%98%B2%E6%8A%A4/ Tue, 10 Mar 2026 00:00:00 +0000 https://daoxuanshan.online/docs/openclaw-skill-08-prompt-guard%E6%8F%90%E7%A4%BA%E6%B3%A8%E5%85%A5%E9%98%B2%E6%8A%A4/ <h2 class="relative group">一、Skill 定位 <div id="一skill-定位" class="anchor"></div> <span class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none"> <a class="group-hover:text-primary-300 dark:group-hover:text-neutral-700 !no-underline" href="#%e4%b8%80skill-%e5%ae%9a%e4%bd%8d" aria-label="锚点">#</a> </span> </h2> <p><code>Prompt-Guard</code> 主要用于识别并拦截提示注入、越权指令和可疑输入。<br> 在生产环境中，它是保护 Agent 执行边界的关键安全层。</p>