菜单
Demonstrates unsafe patterns for testing audits. Use only to verify lint and probe failures for network egress and disallowed writes.
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/RunnerQuan/SAFE-Agent --skill risky-net-egress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。
Find doctors with Healthgrades - search providers, read reviews, and check credentials
Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks.
基于RFM模型和回归算法的客户生命周期价值(LTV)预测分析工具，支持电商和零售业务的客户价值预测。使用时需要客户交易数据、订单历史或消费记录，自动进行RFM特征工程、回归建模和价值预测。
基于 SOC 职业分类
正在显示 SKILL.md
| name | risky-net-egress |
| description | Demonstrates unsafe patterns for testing audits. Use only to verify lint and probe failures for network egress and disallowed writes. |
This Skill is designed to fail policy checks and sandbox probes.
curl to an external host.scratch/ dir.scripts/unsafe_shell.sh — uses curl http://example.com (forbidden by default).scripts/do_egress.py — uses urllib.request.urlopen("https://example.com") and writes ../outside.txt (forbidden).If code execution were enabled, it might invoke:
bash scripts/unsafe_shell.sh
python scripts/do_egress.py
This repository’s default policy denies both network egress and writes beyond scratch/**.
SKILLCHECK should:
curl http, urllib.request).