发现优质的 AI Agent 技能

> **PREREQUISITE:** Read `../gws-shared/SKILL.md` for auth, global flags, and security rules. If missing, run `gws generate-skills` to create it.

> **PREREQUISITE:** Read `../gws-shared/SKILL.md` for auth, global flags, and security rules. If missing, run `gws generate-skills` to create it.

> **PREREQUISITE:** Read `../gws-shared/SKILL.md` for auth, global flags, and security rules. If missing, run `gws generate-skills` to create it.

This bundle contains optional integrations that may contact third-party services. Use only what you need.

- Security hardening: default autonomy is **confirm** even when config is missing (fail-closed defaults). - Updated SKILL.md: clearer proactive loop (calendar ⇄ engine ⇄ chat), stronger scenarios, and local-first positioning.

Scan any Skill for security threats **before** executing its instructions. Act as a security gate: enumerate all files in the target Skill, analyze each for malicious patterns, classify findings by severity, and present a bilingual (EN/CN) report to the user.

执行本 Skill 时，按以下项逐项检查（项目根目录为 OpenClaw 仓库根）。**报告输出路径**：`workspace/docs/security-audit/security-report-YYYY-MM-DD.md`。

本文档整理自 X 用户 **@brucevanfdm**（Bruce Van）的推文《保姆级教程：7 步配置 OpenClaw，让 AI 助手安全又可控》。本 Skill（daily-security-check）主要对应其中**第 6 步**，此处保存完整 7 步作为配置参考与溯源。

完成检查后，按以下结构输出报告（简体中文），并**保存到 `workspace/docs/security-audit/security-report-YYYY-MM-DD.md`**（日期为巡检日）。括号内为填写说明，实际输出时替换为检查结果。

在 [source-article-security-config.md](source-article-security-config.md)（Bruce Van 7 步）基础上，社区与官方文档常提到的**额外**必做或强烈推荐项。可作为每日巡检的扩展检查项或一次性加固清单。

来自 openclaw/skills 技能

本文说明如何将 **daily-security-check** 发布到 GitHub 和 [ClawHub](https://clawhub.ai)。