More screenshots
Works with the AI Agents you already use
Windows AD: Hunt Attack Paths in Active Directory
Hunt Active Directory attack paths with a graph-first methodology — enumerate identity, map paths to domain compromise, prioritize by exploitability, and validate with restraint.
$5
Windows AD: Hunt Attack Paths in Active Directory
Example session with this skill installed
Our AD has grown messy over five years. Map the attack paths to domain compromise and tell us which to close first.
- Read your context and instructions
- Compiled the windows ad
Gate zero passed in writing: authorization from the IT owner, scope naming the single domain and its member servers, and the production database servers explicitly out of scope for validation. Phase 1 enumeration: 214 user accounts, 38 groups, 61 computers, and 4 service accounts — with 17 stale accounts that had not logged in for over a year, and 2 service accounts with passwords older than the domain itself. Phase 2 graph: built from the enumeration, showing every membership, delegation, and admin relationship as an edge — the graph, not the list, is what makes the paths visible. Phase 3 path families: the shortest path to domain compromise ran three hops — a helpdesk account with local admin on a workstation, that workstation's cached credential of a server admin, and the server admin's membership in Domain Admins. Two more paths ran through the stale service accounts. Phase 4 validation with restraint: the first hop was confirmed by checking group membership and local admin rights from the enumeration data alone — no credential was used, no session was hijacked, and the production servers were never touched. Report delivered with the fix order: first, remove the helpdesk account's local admin rights and the stale accounts; second, reset the service account passwords and give them least privilege; third, break the server admin's Domain Admins membership into a separate, audited role. The self-check closed the engagement: enumeration data handled as sensitive, scope respected throughout, and the report handed to the owner with the graph attached.
Connects securely to your tools. The creator never sees your data.
About this skill
Active Directory is a graph, and the shortest path through it to domain compromise is usually three hops long — through an identity nobody is watching. This skill hunts those paths with a graph-first methodology: enumerate identity, build the graph, map the path families, prioritize by exploitability, and validate with restraint. Validation is deliberately restrained: the point is to prove a path is real, not to exploit it, and every validation step stays inside the written scope. Use it only for directories you own or are authorized in writing to test, when the goal is a report that closes paths before an attacker walks them — never for anything else.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
It covers only directories you own or are authorized in writing to test — authorization is gate zero.
Validation proves a path is real using enumeration data; it does not exploit the path or use credentials.
The graph is only as complete as the enumeration — objects the enumeration missed are paths the graph cannot show.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Permissions
No special permissions declared or detected