Sysadmin Assistant Skill

A lot of sysadmin work is running a command you have never run before on a server you cannot afford to break. This skill explains what a command does before you commit, helps diagnose problems from symptom to fix, and acts as a second opinion before you touch something in production.

What the skill does

• Command explanation. Explain what a command does before running it — every flag, what it changes, and what could go wrong.
• Diagnosis. Given symptoms (high load, full disk, failed service, slow response times), produces a diagnostic checklist starting with the cheapest checks.
• Service management. systemd service configuration, enabling/disabling, debugging failed units, journal log analysis.
• User and permission management. User creation, group management, sudo configuration, SSH key setup, and file ownership.
• Performance investigation. CPU, memory, disk I/O, and network — which metrics to check, in what order, and what they mean when they are high.
• Log analysis. Interpreting journal logs, syslog, and application logs — finding the signal in the noise.
• Configuration review. Reviews config files (nginx, sshd, fail2ban, fstab) for correctness and security issues.

How it works

1. Describe the situation. What is the server running? What is the problem or task? What have you already checked?
2. Ask the question. What a command does, how to fix a service, how to investigate a performance issue — anything goes.
3. Explanation or diagnosis. The skill explains clearly, without assuming you know the internals, and without burying the answer in caveats.
4. Step-by-step guidance. For complex tasks, the skill walks through each step with the exact command and what to expect.
5. Safety checks. For destructive or irreversible operations, the skill notes the risk and suggests a verification step first.

How to use it

1. Click ⬇ Download this Claude Skill above.
2. Import through Claude Desktop (Customize → Skills → + → Create skill → Upload a skill) or drop into .claude/commands/. Full walkthrough in the import tutorial.
3. Invoke the skill:

   /sysadmin-assistant
   

   Or ask directly:

   /sysadmin-assistant Ubuntu 22.04. Load average is 45 on a 4-core machine. The app has been running normally for weeks. Started 2 hours ago. Where do I look first?
   

---

Quick-start prompt (no download)

You are a senior Linux sysadmin. Help me with the following.

Before recommending any command:

• Explain what it does and what it changes
• Note any risks or irreversible effects
• Suggest a way to verify the result after running it

For diagnostic tasks, work through the checklist from cheapest/safest checks to more invasive ones. For configuration tasks, show the exact file and line to change, and how to reload the service safely. For anything that could break the server, say so explicitly before I run it.

OS / distro: [Ubuntu 22.04 / Debian / RHEL / etc.] Problem or task: [describe it here] What you have already checked: [if anything]

Tips:

• Load average above the CPU core count is a signal worth investigating, not always a crisis — ask the skill to explain the difference.
• Before changing sshd_config, always keep an existing session open and test in a new terminal window — never lock yourself out.
• Use SysEmperor's fstab Editor to build and validate fstab entries before writing them to disk.
• journalctl -u servicename -n 100 --no-pager is the fastest way to get useful context on a failing service — paste that output when asking for diagnosis help.

⚠ This skill has been tested and optimized for Claude. Results may vary with other AI assistants.