菜单
Apply least-privilege tool allowlisting, identity binding, and human-in-the-loop controls for agent tool calls.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/nexuslinkproductions/yuri-os --skill cyber-securing-agentic-ai-tool-invocation命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | cyber-securing-agentic-ai-tool-invocation |
| description | Apply least-privilege tool allowlisting, identity binding, and human-in-the-loop controls for agent tool calls. |
| hide | true |
Authoritative source: .claude/skills/cyber-securing-agentic-ai-tool-invocation/SKILL.md
Authoritative source SHA-256: 614a2a70ac986273a93dcf7be70fde0c9ac4ea2a2dd6e43bd6924a7a072cdcc1
Source class: cyber-armed
Before acting, read the authoritative source file above completely from beginning to end. If the governed source is absent, run node _SYSTEM/Scripts/skill-recall.mjs --show cyber-securing-agentic-ai-tool-invocation and read its complete verified output. Follow that source as the skill body; this adapter is a non-authoritative metadata-and-pointer projection.