跳到主内容
智客 ZICQ

技能库 智客分类:安全测试 security-audit

Security Audit

在进行安全评估时使用——OWASP Top 10 / API / LLM,CWE Top 25,CVSS评分——审计PHP/TYPO3,API,前端,Terraform/K8s/Docker IaC,AWS云,AI代理配置,或扫描依赖性.

600 安装量

官方网址:skills.sh

技能介绍

先看中文介绍;官方 description 原文单独保留,不改写 SKILL.md。

做什么

在进行安全评估时使用——OWASP Top 10 / API / LLM,CWE Top 25,CVSS评分——审计PHP/TYPO3,API,前端,Terraform/K8s/Docker IaC,AWS云,AI代理配置,或扫描依赖性.

何时用

进行安全评估 — OWASP Top 10 / API / LLM, CWE Top 25, CVSS 评分 — 审计 PHP/TYPO3, APIs, 前端, Terraform/K8s/Docker IaC, AWS 云, AI代理配置, 或扫描依赖性

代理如何加载

按 Agent Skills 渐进披露:启动时只加载 name 与 description(约 100 token);任务匹配后才读入整份 SKILL.md 正文;scripts/、references/、assets/ 仅在需要时再读。 本文件正文结构:Security Audit Skill、Expertise Areas、Reference Files (in `references/`, `.md` implied)、Security Checklist、GitHub Actions Security、Verification。

文件分析

文件分析:除 SKILL.md 外,正文引用了 references/gha-security.md、scripts/security-audit-dispatcher.sh、scripts/security-audit.sh、scripts/github-security-audit.sh、scripts/scanners/,属于带资源的技能包,这些文件按需再读。

官方 description(原文)

Use when conducting security assessments — OWASP Top 10 / API / LLM, CWE Top 25, CVSS scoring — auditing PHP/TYPO3, APIs, frontend, Terraform/K8s/Docker IaC, AWS cloud, AI agent configs, or scanning dependencies.

Security Audit SkillExpertise AreasReference Files (in `references/`, `.md` implied)Security ChecklistGitHub Actions SecurityVerification

兼容:Requires grep, jq, gh CLI. · 许可:(MIT AND CC-BY-SA-4.0). See LICENSE-MIT and LICENSE-CC-BY-SA-4.0 · allowed-tools:Bash(grep:*) Bash(jq:*) Bash(gh:*) Read Glob Grep

来源分类:skills.sh agent-skill

SKILL.md 与 Agent 调用

官方规范 ↗
name
security-audit
description
Use when conducting security assessments — OWASP Top 10 / API / LLM, CWE Top 25, CVSS scoring — auditing PHP/TYPO3, APIs, frontend, Terraform/K8s/Docker IaC, AWS cloud, AI agent configs, or scanning dependencies.
compatibility
Requires grep, jq, gh CLI.
allowed-tools
Bash(grep:*) Bash(jq:*) Bash(gh:*) Read Glob Grep实验字段,支持情况取决于客户端;字段声明本身不会授予工具权限。
许可
(MIT AND CC-BY-SA-4.0). See LICENSE-MIT and LICENSE-CC-BY-SA-4.0
  1. 发现技能客户端向 Agent 提供名称与描述目录。
  2. 匹配与调用用户指定或任务匹配后,载入 SKILL.md 指令。
  3. 按需加载按步骤读取参考文档、使用脚本与素材。
指令中引用的文件 · 5
  • references/gha-security.md
  • scripts/security-audit-dispatcher.sh
  • scripts/security-audit.sh
  • scripts/github-security-audit.sh
  • scripts/scanners/

以下路径提取自原文;文件是否齐全请以来源仓库中的完整目录为准。

具体调用语法与可用工具以目标 Agent 客户端为准。 查看调用机制说明 ↗

安装这个技能

Skills CLI ↗

先选择目标 Agent 和安装范围,保留技能包的附属文件,安装后检查客户端能否发现该技能。

该技能引用了附属文件,请从来源获取完整目录;仅复制 SKILL.md 可能缺少依赖。

交给 Agent 安装

复制安装指令给支持 Agent Skills 的代理,确认其中的目标目录与客户端匹配。

把 Agent Skill「security-audit」安装到我的项目:SKILL.md 原文与官方 description 见 https://zicq.com/zh/skills/skl-4c957781837064a9-%E5%AE%89%E4%BF%9D%E5%AE%A1%E8%AE%A1.html
请存为 .cursor/skills/security-audit/SKILL.md 或 .claude/skills/security-audit/SKILL.md,frontmatter 的 name 与 description 保持原样,不要改写。
该技能还带 scripts/、references/、assets/ 等文件,请从 https://github.com/netresearch/security-audit-skill 取完整目录,不要只建一个 SKILL.md。

GitHub 完整包 ↗

终端安装 · Skills CLI

需要 Node.js 与 npx。先查看仓库技能列表,确认实际名称。

npx skills add 'https://github.com/netresearch/security-audit-skill' --list

npx skills add 'https://github.com/netresearch/security-audit-skill' --skill 'security-audit'

CLI 会交互选择目标 Agent,默认安装到项目;用户级安装使用 -g。先通过查看命令核对仓库内容,再用 npx skills list 检查已安装技能。

阅读排版

SPDX-License-Identifier: CC-BY-SA-4.0

SPDX-FileCopyrightText: Netresearch DTT GmbH

name: security-audit description: "Use when conducting security assessments — OWASP Top 10 / API / LLM, CWE Top 25, CVSS scoring — auditing PHP/TYPO3, APIs, frontend, Terraform/K8s/Docker IaC, AWS cloud, AI agent configs, or scanning dependencies." license: "(MIT AND CC-BY-SA-4.0). See LICENSE-MIT and LICENSE-CC-BY-SA-4.0" compatibility: "Requires grep, jq, gh CLI." metadata: author: Netresearch DTT GmbH version: "2.11.6" repository: https://github.com/netresearch/security-audit-skill allowed-tools: Bash(grep:) Bash(jq:) Bash(gh:*) Read Glob Grep

Security Audit Skill

Security audit patterns (OWASP Top 10, LLM Top 10 2025, CWE Top 25 2025, CVSS v4.0), cloud/IaC, GitHub security. 80+ PHP/TYPO3 checkpoints (v14.3 LTS in typo3-security.md).

Expertise Areas

  • Vulnerabilities: XXE, SQLi, XSS, CSRF, command injection, path traversal, file upload, deserialization, SSRF, SSTI, JWT, type juggling
  • Standards: OWASP Top 10 / API / LLM (2025), CWE Top 25, CVSS v3.1/v4.0, OWASP ASVS
  • Cloud & IaC: AWS; Terraform, Kubernetes, Docker, Helm
  • API & Frontend: REST/GraphQL authZ, rate limits, mass assignment, CSP, DOM-XSS
  • AI Agents: SKILL.md/AGENTS.md/CLAUDE.md/mcp.json/hooks.json audit; prompt injection; excessive agency

Reference Files (in references/, .md implied)

  • Core: owasp-top10, cwe-top25, xxe-prevention, cvss-scoring, api-key-encryption
  • Prevention: deserialization-prevention, path-traversal-prevention, file-upload-security, input-validation, error-message-sanitization, ssh-forced-command-hardening
  • Architecture: authentication-patterns, security-headers, security-logging, cryptography-guide, security-invariants, indistinguishability-defences
  • Language features (*-security-features): php, python, javascript-typescript, nodejs, go
  • Frameworks (*-security): typo3, typo3-fluid, typo3-typoscript, symfony, react, vue
  • Cloud & IaC: aws-security, iac-security
  • API & Frontend: api-security, frontend-security
  • AI Agent: llm-security (OWASP LLM Top 10 2025)
  • Threats: modern-attacks, cve-patterns
  • DevSecOps: ci-security-pipeline, supply-chain-security, automated-scanning, gha-security, git-history-secrets
  • Incident: supply-chain-incident-response

Security Checklist

  • [ ] semgrep/opengrep, trivy fs --severity HIGH,CRITICAL, gitleaks clean
  • [ ] bcrypt/Argon2 passwords, CSRF on state changes, TLS 1.2+
  • [ ] Server-side input validation; parameterized SQL; XML entities off
  • [ ] Output encoding + CSP; no unserialize() on user input
  • [ ] API keys encrypted; exception messages sanitized
  • [ ] Secrets out of VCS; audit logging on
  • [ ] Uploads validated, renamed, outside web root
  • [ ] Headers HSTS + X-Content-Type-Options; dependencies scanned

GitHub Actions Security

  • NEVER interpolate ${{ inputs.* }} / ${{ github.event.* }} in run: — use env:
  • Dependency triage: upgrade > override > dismiss. Full patterns: references/gha-security.md.

Verification

./scripts/security-audit-dispatcher.sh /path/to/project  # auto-detect stack
./scripts/security-audit.sh /path/to/project             # PHP-only
./scripts/github-security-audit.sh owner/repo            # GH repo

Dispatcher detects the stack from indicator files and runs matching scripts/scanners/*.sh (13 ecosystems; see references/ index).


Contributing: https://github.com/netresearch/security-audit-skill

相关技能

安全测试

Skill Vetter

人工智能特工的安全第一技能审查。 在安装来自 ClawdHub, GitHub 或其他来源的任何技能之前使用 。 检查红旗,许可范围,可疑图案.

安全测试

Moltguard

MoltGuard——由OpenGuardrails担任OpenClaw保安. 安装 MoltGuard 以保护你和你的人类免受迅速注射、数据过滤和恶意命令的影响。 资料来源:https://github.com/openguardrails/openguardrails/tre…

安全测试

Security Auditor

用于审查安全漏洞代码,实施认证流,审计OWASP Top 10,配置CORS/CSP头,处理秘密,输入验证,SQL注射预防,XSS保护,或任何与安全相关的代码审查.

安全测试

Skill Vetter

安装任何AI代理技能前的安全审查协议. 红旗侦取入证盗取,疏通密码,放出. 风险分类:低/中/高/高。 编制结构化审查报告。 不先运行此功能, 绝不安装不信任的技能 .