📦 skill-governance — 技能治理核心
v2.1.0OpenClaw Cognitive Operating & Skill Governance Kernel，为 AI 技能提供统一注册、版本控制、权限与生命周期管理，确保多技能协同安全可控。
0· 390·0 当前·0 累计
by 下载技能包
最后更新
2026/4/22
安全扫描
OpenClaw
可疑
medium confidenceNULL
评估建议
This skill is conceptually coherent with a governance kernel, but it contains ambiguous operational directives that could change files, archive or delete other skills, and push summaries externally without specifying where. Before installing: 1) Ask the author to clarify exact filesystem paths and required permissions for /memory and archived_skills/, and which service/endpoints (and credentials) are used for 'external synchronization'. 2) Confirm which bundles the skill may mount/unmount and ge...详细分析 ▾
✓ 用途与能力
The name/description as a 'cognitive operating & skill governance kernel' matches the SKILL.md: it defines perception, decision, mounting, lifecycle, and closure protocols. There are no unexpected environment variables, binaries, or installs requested that contradict its stated purpose.
⚠ 指令范围
The instructions mandate behaviors that involve system state and other skills: automatic 'mount/unmount' of bundles, writing forced archives to /memory/YYYY-MM-DD-task.md, moving skills to archived_skills/, and generating summaries for 'external synchronization'. The SKILL.md does not specify where /memory or archived_skills/ live, what API/endpoints should be used for external sync, or what authorization is needed. Those gaps create a risk that the agent will read/write files or transmit data outside expected boundaries or trigger other skills unexpectedly.
✓ 安装机制
No install spec and no code files — instruction-only — so nothing is downloaded or written at install time. This is lower risk from an install-mechanism perspective.
ℹ 凭证需求
The skill requests no environment variables or credentials (proportionate). However, it references filesystem locations and lifecycle operations that imply write/delete privileges over skill storage areas even though no config paths were declared in the manifest; that mismatch should be clarified.
⚠ 持久化与权限
always:false and model invocation allowed (normal). But the protocol includes lifecycle actions that move and mark other skills (archived_skills/, deletion candidates) and requires sending notifications before deletion. Those are operations that modify other skills' state or system-wide skill storage; the skill does not declare these config paths or required permissions, which is a privilege/footprint mismatch and a potential control risk.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv2.1.02026/3/2
NULL
● 可疑
安装命令
点击复制官方npx clawhub@latest install skill-governance
镜像加速npx clawhub@latest install skill-governance --registry https://cn.longxiaskill.com