How to use the AD Certificate Abuse Playbook agent skill

What does this skill make the agent do?

The skill is a reference playbook. It tells the agent how to structure its analysis of an AD CS setup, and how to keep the evidence organised as it goes. It is not a tool that scans or connects to anything, and the files it ships do not include executable scripts.

It is designed as a downstream step. It only activates after the parent skill ctf-sandbox-orchestrator has set up the sandbox assumptions and routed the task here.

• Identify the certificate authority (CA), the template, the enrolling principal and the accepting service first.
• Separate whether a template can be enrolled in from whether a certificate is accepted for authentication or privilege.
• Record EKUs, subject or SAN controls, issuance requirements, enrollment rights and mapping behaviour in compact blocks.
• Tie an issued certificate to one accepted path, such as PKINIT, Schannel, LDAPS or WinRM.
• Keep the smallest issuance-to-acceptance chain that explains the result, with serials, subjects, SANs, EKUs and validity recorded.

What do you need before installing it?

The skill will not do its job on its own. The source states it should not be used without its parent skill, so install and activate ctf-sandbox-orchestrator first.

You also need an agent runtime that reads SKILL.md-format skills. The source lists no other requirements.

• The parent skill ctf-sandbox-orchestrator installed and active.
• A SKILL.md-compatible agent runtime.
• No scripts run and no network access is used, according to the listing.

How do you install the skill?

Download the ZIP for this skill from this page and unzip it. Then place the skill folder in one of the locations below. Check that SKILL.md sits directly inside the folder you unzipped, alongside references/ad-certificate-abuse.md and agents/openai.yaml.

If you use the Claude apps instead of a folder, you can upload the skill under Settings → Capabilities → Skills.

• For personal use across projects: unzip into ~/.claude/skills/.
• For a single project: unzip into the project's .claude/skills/ folder.
• In the Claude apps: upload it under Settings → Capabilities → Skills.

How do you trigger the skill with a natural request?

You do not need a special command. Ask in plain language about certificate topics. The skill is written to activate on requests about certificate templates, enrollment agents, EKUs, SAN or subject controls, smartcard or PKINIT logon, CA policy, and how an issued certificate becomes accepted privilege.

Name the parent skill in your first message so the routing is clear. The skill also replies in Simplified Chinese by default, so ask for English if you want it.

• Explain how this certificate template could end up granting logon privilege in my authorized lab.
• Map the CA, template and enrolling principal for this challenge before we look at the certificate details.
• Which EKUs and SAN settings on this template matter for smartcard logon?
• Turn what we found into a compact evidence record: serial, subject, SAN, EKU, validity and issuer.

What are typical examples of use?

The most common use is a CTF challenge that involves AD CS. The skill helps the agent work out which certificate step is the decisive one, instead of stopping at a template setting. The source also notes that a security team reviewing its own AD CS template configuration for misconfigurations is an intended audience.

Because the skill emphasises evidence, it is also useful when you need to write up a finding clearly, with the template side, the certificate side and the acceptance side kept together.

What safety limits should you know?

The skill assumes a sanctioned sandbox. Its guidance on abuse chains is harmful outside authorised testing, so do not use it against production or third-party networks.

The skill is also oriented toward exploitation paths rather than hardening. If you need general Windows administration or certificate setup help, use a different resource.

Because the skill runs no scripts and uses no network, it adds no tooling of its own. Whatever the agent does against a lab comes from your runtime and your instructions, so keep that work inside the sandbox you were given.

What tips help you get better results?

• Give the agent the CA name, template name and enrolling principal up front.
• Ask it to keep the template, certificate and acceptance evidence together.
• Ask it to separate certificate issuance from the separate step where privilege is granted.
• If your question is really about delegation or ticket transformation after PKINIT, the source says to switch to a Kerberos-focused skill instead.