⏳ This skill is pending AI review.

Scores will appear once the review pipeline completes.

version unknown

ad-attack

@pale-knight⭐ 49 stars

Active Directory exploitation after domain credentials exist: Kerberos (AS-REP/Kerberoast including cracking the ticket then using the account), delegation, NTLM coercion/relay, lateral movement, ACL abuse, ADCS ESC1-ESC17 and CVE paths, dMSA/BadSuccessor, Kerberos reflection, identity confusion, management-plane, domain trust, and domain persistence. Own the current AD chain through crack-and-use to DA, equivalent domain control, or the targeted host SYSTEM. Do not stop after requesting TGS. Host C2 belongs to /post. Operator chooses next modules.

Choose how to use this skill

You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.

1. Native installer

This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.

Do not guess an installer command or replace an existing version without reviewing the diff.

2. Complete package recommended

Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.

No complete ProSkills package is published for this listing yet.

3. Prompt-only

Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.

Need only the instruction file?

Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.

No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.

—/10

// RATINGS

⭐GitHub Stars
⭐⭐ 49 on GitHubGitHub ↗

Growing

🟢ProSkills Score
—
📍

Not yet listed on ClawHub or SkillsMP

// README

关于项目

辅助红队工作,操作者通过/module选择模块,AI将取得的成果记录在notes.md中供操作者研判并建议下一模块,并在自身无法做到需要操作者辅助,例如:监听/sudo/输密码/Permission denied等情况时停止并等待操作者完成

本项目面向红队工作 / 靶场 / HTB 等场景,提供半自动红队工作流。

做成半自动是因为真实红队中,合格的渗透人员必须根据现场情况做研判并选择下一步,而不是把整条链路交给 AI。AI 可能执行不该执行的动作或者漏掉重要的信息,并无法承担相应的责任和带来的后果。所以也没有限制目标授权的范围,这也应该交由渗透人员负责。

此项目集成了绝大部分技能模块、最近工作中的高成功率利用链,以及 2025–2026 较新的技术路径,并会持续更新。

为了缩短上下文,节约token。此项目将每一模块的重要发现和成果记录在notes.md中,当前模块结束时可由人工/clear清除上下文。在每一个模块开始时会自动读取notes.md中数据,以保证工作延续。并且将每一模块的详细操作流程放入references中,只在遇到时才会读取对应的references/<file>.md。

项目架构

基本架构

18 个模块,每个模块为SKILL.md(判断基本方向)和一个references/(详细操作流程)。

CLAUDE.md                  # 全局规则,放进项目目录使用
skills/
├── recon/                     # 通用信息收集
│   ├── SKILL.md
│   └── references/     (11)
├── service-attack/            # 服务漏洞利用
│   ├── SKILL.md
│   └── references/     (13)
├── web-recon/                 # Web攻击面测绘
│   ├── SKILL.md
│   └── references/     (5)
├── web-attack/                # Web漏洞利用
│   ├── SKILL.md
│   └── references/     (24)
├── ad-recon/                  # AD枚举
│   ├── SKILL.md
│   └── references/     (5)
├── ad-attack/                 # AD攻击利用
│   ├── SKILL.md
│   └── references/     (9)
├── cloud-recon/               # AWS/Azure/GCP/阿里云枚举
│   ├── SKILL.md
│   └── references/     (7)
├── cloud-attack/              # 云控制面利用
│   ├── SKILL.md
│   └── references/     (9)
├── k8s/                       # 容器与集群
│   ├── SKILL.md
│   └── references/     (6)
├── cicd/                      # CI/CD与供应链
│   ├── SKILL.md
│   └── references/     (12)
├── phishing/                  # 钓鱼和客户端攻击
│   ├── SKILL.md
│   └── references/     (8)
├── privesc-win/               # Windows提权
│   ├── SKILL.md
│   └── references/     (5)
├── privesc-linux/             # Linux提权
│   ├── SKILL.md
│   └── references/     (5)
├── creds/                     # 凭据攻击
│   ├── SKILL.md
│   └── references/     (7)
├── post/                      # 后渗透利用与C2
│   ├── SKILL.md
│   └── references/     (7)
├── shell/                     # shell稳定化
│   ├── SKILL.md
│   └── references/     (6)
├── tunnel/                    # 内网穿透
│   ├── SKILL.md
│   └── references/     (7)
├── edr-bypass/                # AV/EDR规避
│   ├── SKILL.md
│   └── references/     (12)
├── shared/                    # 跨模块共享
│   ├── modules.yaml           # 模块目录
│   ├── cve-enrichment.md      # 全Skill漏洞情报入口
│   ├── tools.md               # 工具注册表
│   └── wordlists.md           # 字典路径
└── bin/
    ├── modules.py             # tail / list / show / check,模块控制和收尾
    └── notes.py               # init / validate,操作记录文件notes.md

references/内共 158 份操作文档,按SKILL.md 判断选择加载,不计入常驻上下文。

模块介绍

场景入口类型功能成功条件默认下一步
通用信息收集/reconrecon资产扩展、主机与端口发现、服务版本识别、只读枚举、CVE 候选研判服务地图 + CVE候选/web-recon /ad-recon /service-attack /phishing
服务漏洞利用/service-attackattack数据库、文件与远程访问、消息队列、DNS、网络设备与BMC等利用链对应服务的foothold或shell/shell /privesc-win /privesc-linux /creds /post
Web攻击面测绘/web-reconrecon指纹、路径与 API 发现、JS/sourcemap、代理缓存边界、WAF、CMSWeb攻击面卡片/web-attack
Web漏洞利用/web-attackattack注入、上传、LFI、SSRF/XXE/SSTI、反序列化、JWT/SAML、走私、WAF 绕过等利用链foothold、shell或等价OS执行/shell /privesc-win /privesc-linux /creds /post
域枚举/ad-reconrecon获取用户组机器、ACL/委派、ADCS、LAPS、BloodHound、信任图等域路径与 ACL 卡片/ad-attack /creds
域攻击利用/ad-attackattackKerberos、委派、强制认证与中继、ACL 滥用、ADCS ESC 全系、dMSA、横向等攻击DA、等价域控制,或目标主机SYSTEM/ad-recon /creds /post /privesc-win
云身份枚举/cloud-reconrecon获取AWS/Azure/GCP/阿里云的身份、IAM、信任、资源、元数据等云身份、权限、信任、资源图/cloud-attack /k8s
云控制面利用/cloud-attackattackIAM 提权、身份假冒、跨账户信任、serverless、云原生持久化更高云身份、账户控制,或可下发OS执行的 compute/cloud-recon /k8s /post /ad-recon
容器与集群/k8sattackRBAC 滥用、secrets、kubelet/etcd、容器逃逸、云身份绑定、集群持久化等cluster-admin、node root,或可用的云身份/cloud-recon /post /privesc-linux /creds
CI/CD 与供应链/cicdattackJenkins、GitHub Actions、GitLab/ADO、runner、依赖混淆、registry 投毒、OIDC等利用runner shell、部署控制,或可独立使用的新身份/cloud-recon /privesc-linux /privesc-win /post
钓鱼与客户端攻击/phishingattackClickFix/FileFix、AiTM 会话、device code/OAuth、helpdesk 社工、文件投递等攻击host shell或foothold/shell /privesc-win /privesc-linux /creds /post
Windows 提权/privesc-winattackPotato家族、令牌特权、服务/DLL/计划任务、UAC、内核LPE等方式Administrator(High IL)或NT AUTHORITY\SYSTEM/creds /post /ad-recon /tunnel
Linux 提权/privesc-linuxattacksudo/GTFOBins、polkit、SUID/capabilities、systemd、危险组、内核LPE等方式uid=0 root shell/creds /post /k8s /cloud-recon
凭据攻击/credsfactorysecret发现、离线破解、按策略喷洒、NetNTLM捕获与SMB relay、系统凭据收割经校验的可用凭据/ad-recon /cloud-recon /service-attack /privesc-win /privesc-linux /post
后渗透与C2/postpost主机画像、host-native持久化、Sliver C2、定向收集外带、清理所选目标已验证(回连/persist/loot/已还原)/tunnel /creds /ad-recon /cloud-recon
shell 稳定化/shellsupport回连引导、Linux PTY、Windows ConPTY、监听管理、会话恢复、文件传输可操作稳定会话/privesc-win /privesc-linux /creds /post
内网穿透/tunnelsupportligolo-ng、chisel、GOST、原生转发、Dev Tunnels、多跳、备用传输目标网段可达/recon /service-attack /ad-recon
AV/EDR规避/edr-bypassinterceptor将被AV/EDR/AMSI/WDAC/PPL/内存/内核遥测的拦截变为可执行被拦动作可执行回原模块

recon 模块信息收集以确认攻击面,attack 模块执行攻击链获取对应控制权,support 模块不推进攻击链只稳固当前shell,factory 模块仅有/creds 只校验凭据,post 模块仅有/post只做收尾工作,interceptor 模块仅有/edr-bypass只在被拦截时执行,在绕过拦截之前切回原模块继续执行。

使用说明

前置依赖

  • Kali Linux — 推荐环境,工具链默认按Kali路径和 shared/tools.md 内容安装
  • Python 3.x — 运行 bin/modules.py、bin/notes.py
  • Claude Code — 本 Skill 面向 Claude Code 设计,模块经 /slash 命令调用

目录层级

工作目录

~/ops/<target>/
├── CLAUDE.md                  # 工作规则
├── notes.md                   # 工作记录
└── scans/  loot/  scripts/    # 工作产出

可以自己建立或者让AI自动创建。

工具目录

~/tools/
├── recon/
├── web/
├── ad/
...
├── c2/
├── edr/
└── shell/

详情见shared/tools.md。

安装

git clone https://github.com/pale-knight/redteam-skill.git

​> 把18个模块、shared/和bin/移动到~/.claude/skills/skill的全局目录,CLAUDE.md放入每个项目目录。

开始

mkdir -p ~/ops/<target> && cd ~/ops/<target>
cp CLAUDE.md ./CLAUDE.md

创建notes.md

python ~/.claude/skills/bin/notes.py init

单次工作流程

操作者   /<module>
AI      Read ./notes.md
        阅读SKILL.md
        走到对应流程,才读取对应的references/<file>.md
        做到本模块成功条件
        需要操作者协助时 → 立刻停,说明

// HOW IT'S BUILT

KEY FILES

skills/ad-attack/SKILL.mdREADME.md

// REPO STATS

49 stars