by 安全扫描
OpenClaw
安全
high confidenceNULL
评估建议
This skill is a high-level, instruction-only audit workflow — it won't install code or request secrets. Before using it, decide which documents and logs the agent should access (regulations, incident history, configs), redact or withhold sensitive data as needed, and provide only the minimal inputs required. Because the instructions are broad, restrict agent access scope when running (explicit input files, limited folders) and review produced config diffs before applying them to production. If y...详细分析 ▾
✓ 用途与能力
Name, description, and included workflow files all describe a regulation-audit process; there are no unexpected binaries, credentials, or installs required that would be disproportionate to this purpose.
ℹ 指令范围
SKILL.md defines the expected audit steps and outputs and references the included workflow file. The instructions are intentionally high-level (e.g., "Collect current regulation sources") which is appropriate for a template-style skill but grants broad discretion about what data sources to access; users should be aware that the agent will need access to whatever regulation, incident, and config artifacts you provide.
✓ 安装机制
No install spec and no code files — instruction-only — so nothing will be written to disk or downloaded during install.
✓ 凭证需求
The skill declares no required environment variables, credentials, or config paths; that aligns with its stated purpose as a guidance/workflow skill.
✓ 持久化与权限
always is false and the skill does not request elevated or persistent privileges. Autonomous invocation is allowed by default but that is normal; nothing in the skill asks to alter other skills or system-wide settings.
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
latestv0.1.02026/2/22
NULL
● 可疑
安装命令
点击复制官方npx clawhub@latest install presale-regulation-auditor
镜像加速npx clawhub@latest install presale-regulation-auditor --registry https://cn.longxiaskill.com