security-audit
每次发现都会转成可复用的检测规则，能随事件持续增强覆盖率。
需书面授权并可能执行敏感扫描或提交检测规则，存在误用或误报风险。
对 WordPress 进行加固与可复用检测规则的自动学习型安全审计与事件响应。
面向防御的 WordPress 安全评估与事件响应：做加固、检测、遏制和修复，并把每次发现转成可复用的检测规则（如 semgrep、YARA）。适合需要长期提升检测覆盖或在事故后快速形成自动化响应的场景。它会在扫描前加载历史检测规则、在事后把新发现写成规则提交，从而让检测能力随时间不断增长。
▸ 展开 SKILL.md 英文原文
WordPress security posture audit + incident triage/response that LEARNS FROM EVERY FINDING. Auto-learning by design AND incident-learning: it loads its accumulated detection rules (YARA/semgrep) before scanning, and turns each new incident/finding into a reusable detection signature it commits — so it catches more next time. A defensive orchestrator over Trail of Bits (`static-analysis`, `semgrep-rule-creator`, `yara-authoring`, `supply-chain-risk-auditor`), the `ffuf` skill, `wordpress-penetration-testing`, and E-Lego's Elementor security-audit branch.
帮我安装这个 skill：https://raw.githubusercontent.com/dxxx/Bot-OS/main/skills/security-audit/SKILL.mdcurl -fsSL "https://raw.githubusercontent.com/dxxx/Bot-OS/main/skills/security-audit/SKILL.md"# WordPress Security Audit & Incident Learning > Defensive security: **harden → detect → contain → remediate → LEARN.** Every real finding > becomes a durable lesson *and* a reusable detection rule, so the skill's coverage grows with > each incident it sees. Assumes you own or are authorized to secure the target. ## ⛔ Safety gate (first, every time) - [ ] **Authorized** to audit/remediate this site (you own it or have written permission). - [ ] **Back up first** (DB + files + config) before any state-changing remediation; prefer a snapshot. - [ ] **Contain, don't destroy:** quarantine suspicious files (don't blind-delete — preserve evidence + rollback). - [ ] **Remediation is live-to-exte