Standard operating procedure for solving Windows and Active Directory TryHackMe boxes. Use when the target is Windows/domain-joined (SMB/LDAP/Kerberos/WinRM ports open, or the room mentions AD/domain), when asked "what next" on a Windows room, or when a Linux-oriented approach isn't fitting. Guides the recon → AD enum → foothold → Kerberos → privesc → domain flow.
Scanned 9/6/2026
Install to Claude Code
npx -y skills add pashki975/thm-claude-kit --skill windows-ad-methodology --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Windows Ad Methodology?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/pashki975-windows-ad-methodology)More formats (shields.io, HTML) on the badges page.
---
name: windows-ad-methodology
description: Standard operating procedure for solving Windows and Active Directory TryHackMe boxes. Use when the target is Windows/domain-joined (SMB/LDAP/Kerberos/WinRM ports open, or the room mentions AD/domain), when asked "what next" on a Windows room, or when a Linux-oriented approach isn't fitting. Guides the recon → AD enum → foothold → Kerberos → privesc → domain flow.
---
# Windows / Active Directory Methodology
For readers who live in Linux: Windows CTF is a different surface. Instead of web/SSH →
SUID/sudo, the game is SMB/LDAP/Kerberos → tokens/ACLs/tickets. Follow this flow.
## Recognizing a Windows box
Open 135/139/445 (SMB), 3389 (RDP), 5985 (WinRM), and especially 88 (Kerberos) +
389/636 (LDAP) → domain-joined AD. Kerberos (88) present = treat it as full AD.
## 0. Setup
- /vpn-check the target
- Add the DC's FQDN to /etc/hosts (Kerberos needs name resolution)
- Watch clock skew: sync to the DC if Kerberos tooling errors (`sudo rdate -n <DC-IP>`)
## 1. Recon
- /recon as usual. Note the domain name, DC hostname, OS build.
- Consider /enum-udp too — SNMP can leak users/processes on Windows hosts.
## 2. AD / SMB enumeration
- /enum-ad — null/guest sessions, shares, RPC, LDAP, build a user list
- Harvest usernames from RID cycling, shares, LDAP. Save to users.txt.
## 3. Get a foothold credential (try in order)
- AS-REP roast the user list (/kerberos) — no creds needed
- Password spray weak/found passwords across users (careful of lockout)
- Anonymous/guest readable shares — look for creds, configs, scripts
- Any web app on the box → normal web enum (/enum-web)
- Any database on the box → /db-enum (MSSQL especially — xp_cmdshell)
## 4. Situational awareness (once you have ANY valid cred)
- Run BloodHound collection — let the ad-attack-advisor agent read the results
- Validate the cred everywhere: `nxc smb <IP/range> -u <u> -p <p>` (look for Pwn3d!)
## 5. Escalate within the domain
- Kerberoast service accounts (/kerberos), crack offline
- Abuse ACLs / delegation / DCSync per BloodHound (ad-attack-advisor agent)
- Check ADCS (certipy) if certificate services are present
## 6. Get a shell
- /win-shell — evil-winrm, psexec/wmiexec, or RDP depending on the cred
- Immediately run `whoami /priv` and `whoami /groups`
## 7. Local privesc (if not already SYSTEM/admin)
- /win-privesc — token privileges first (SeImpersonate → Potato is the classic),
then services, stored creds, kernel last
- Analyze enum with the win-privesc-advisor agent
## 8. Pivot (if the domain has more hosts to reach)
- From the compromised host, tunnel to internal targets (/tunnel: ligolo-ng / chisel)
- Reuse domain creds across the internal range — one cred often unlocks many hosts
- Record each internal host in notes.md and continue this methodology against it
## 9. Domain dominance & loot
- DCSync krbtgt if you reach it — note the domain is fully compromised
- Record all flags (user.txt, root.txt/admin) and creds in notes.md
- report-writer agent for the writeup
## Rules
- Stay strictly on the assigned target IP/range
- Most Windows CTF blockers are: missed enum, /etc/hosts, or clock skew — check those first
- Prefer credentialed remote tools (evil-winrm) over noisy reverse shells when you can
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!