Guide · Agent skill
How to use the Offensive Lateral Movement agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The offensive-lateral-movement skill is a reference playbook that gives an AI agent detailed lateral movement tradecraft for authorized red team and internal penetration test work: credential reuse, NTLM relay, remote execution, session hijacking and network pivoting. This guide explains what the skill makes the agent do, how to install it, how to trigger it with natural requests, and what safety limits apply. It is written for readers who have not used agent skills before, and it assumes you already have written authorization for the target network.
Offensive Lateral MovementSnailSploit/Claude-Red · MIT Download the skill Quick start
- 1Confirm authorization. Make sure you have written authorization for the exact network and hosts you will test. Do not start without it.
- 2Download the ZIP. Get the offensive-lateral-movement skill as a ZIP from the skill page.
- 3Unzip into a skills folder. Place the folder in ~/.claude/skills/ for personal use or in a project's .claude/skills/ folder, or upload the ZIP under Settings → Capabilities → Skills in the Claude apps.
- 4Install the attack tooling. Obtain the tools the skill relies on, such as mimikatz, netexec, impacket and evil-winrm, from their own sources. The skill does not install them.
- 5Describe the task in plain language. Ask a request that names lateral movement, the protocol or the credential you hold, and the in-scope targets.
- 6Check every proposed command. Review target addresses, accounts and domains against your engagement rules before running anything.
- 7Record each pivot. Log every host you move to so the findings can go into the engagement report.
What the skill makes the agent do
Agent skills are folders of instructions that an AI agent loads when a task matches them. This one is a single SKILL.md file. It describes the step that comes after initial access on a Windows network: moving from one compromised machine to others using credentials or tickets you already hold.
The source covers pass-the-hash, pass-the-ticket and overpass-the-hash, NTLM relay with coercion tools such as PetitPotam, DFSCoerce and PrinterBug, remote execution over WMI, WinRM, DCOM and PsExec, RDP session hijacking, and tunneling with chisel, ligolo-ng, SSH tunnels and SOCKS proxies. It also includes command sequences for mimikatz, crackmapexec/netexec, the impacket suite and evil-winrm, plus notes on OPSEC, detection and MITRE ATT&CK mapping (T1021 and T1550).
Install the skill
Download the skill as a ZIP from the skill page, then unzip it. For a personal install, place the skill folder in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder.
In the Claude apps, you can instead upload the ZIP under Settings → Capabilities → Skills.
- Personal install: ~/.claude/skills/offensive-lateral-movement/
- Project install: .claude/skills/offensive-lateral-movement/ inside your project
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
Trigger the skill with natural requests
You do not need a special command. Describe the task in plain language that mentions lateral movement, the protocol or the credential you hold. The agent should load the skill when your request matches its description.
Give the agent the context it needs: the authorized scope, which credentials or hashes you hold, and which hosts or ports you have found.
- Given this NTLM hash for admin on the 10.10.10.0/24 subnet, which hosts accept it for SMB?
- Which hosts in this subnet do not require SMB signing, so I can plan an ntlmrelayx run?
- I have a shell on ws01. Show me the options for running commands on dc01 over WinRM or WMI.
- I need to reach a segmented network through this pivot host. Outline a tunnel setup with chisel or ligolo-ng.
Read the outputs with care
The skill supplies command syntax, not judgment about your scope. Check every target address, account and domain against your engagement rules before you run anything the agent proposes.
Keep a record of each pivot. The skill's own workflow says to document each pivot for the engagement report.
Safety notes: scripts and network access
This skill runs no scripts of its own. It does use the network: the commands it describes, once you run them, enumerate hosts, authenticate to services and can dump credentials. The source lists the tools the commands rely on, including mimikatz, netexec, impacket, evil-winrm, Responder and tunneling tools, and you must obtain and install them yourself.
Only run these against systems you are authorized to test. The skill itself says to apply these techniques in authorized engagements only. Credential spraying can cause account lockouts, and credential dumping can disrupt production systems, so avoid production environments where an agent could cause harm.
- Requires valid credentials or hashes from an authorized engagement
- Requires a Windows target environment and the attack tooling named above
- The skill runs no scripts; it uses network access
- Use the source's --no-bruteforce option with crackmapexec when testing multiple users against multiple hashes, to reduce lockout risk
Tips for getting useful output
Name the protocol you want. The skill covers several remote execution options, and each leaves a different footprint. Ask for the stealthier option explicitly if that matters to your engagement.
Ask for the detection view when you need it. The skill includes blue-team guidance, such as the Windows event IDs 4768, 4769 and 7045 that these techniques can generate.
- Say which credential type you hold: plaintext password, NT hash, or Kerberos ticket
- Ask for OPSEC notes alongside commands
- Ask for the defender perspective to see which logs an action creates
What people use it for
- Testing whether a captured NT hash is reused across a subnet during an internal penetration test
- Checking which hosts lack SMB signing before planning an ntlmrelayx run in an authorized engagement
- Choosing between wmiexec, psexec and evil-winrm for remote command execution with a known credential
- Planning a tunnel through a pivot host with chisel or ligolo-ng to reach a segmented network
- Mapping which Windows events (such as 4768, 4769 and 7045) an assumed-breach exercise would generate, for a blue team review
Frequently asked questions
How do I install the offensive lateral movement skill?
Download the skill as a ZIP from the skill page and unzip it. Place the folder in ~/.claude/skills/ for personal use or in a project's .claude/skills/ folder. In the Claude apps, you can instead upload the ZIP under Settings → Capabilities → Skills.
Is the offensive lateral movement skill free?
The skill is published under the MIT licence, as listed on its page. Check the repository licence file for the full terms before reuse.
What do I need to use this skill?
You need a Windows target environment, the attack tools the skill names (such as mimikatz, netexec or crackmapexec, impacket, evil-winrm, Responder and tunneling tools), and valid credentials or hashes from an authorized engagement. The skill does not explain the underlying Active Directory concepts.
Does the skill run scripts on my machine?
No. The skill's listing says it runs no scripts. It does use the network, because the commands it describes enumerate hosts and authenticate to services once you run them yourself.
Can a beginner use this skill to learn penetration testing?
Not well. The skill assumes deep Active Directory knowledge and does not explain the underlying security concepts. Beginners should learn the basics first and only use this as a command reference during authorized work.
How does it compare with running the tools by hand?
The skill gives an agent a structured playbook with command syntax, OPSEC notes and detection guidance in one place, so you spend less time looking up flags. It does not replace knowing what each technique does, and you still have to verify every command against your scope.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
This skill suits professional penetration testers on authorized internal or assumed-breach engagements who want a quick reference for lateral movement command syntax, and blue-team engineers who want to see which Windows events these techniques produce. It is a poor fit for anyone without written authorization, for beginners who need the underlying concepts explained, or for production environments where credential dumping or lockouts could cause disruption.
Download the skill