Claude Skill

sysops-administrator

Administer Linux, macOS, and Windows servers and VMs - services, disks, memory, processes, network, certificates, backups, patches. Whitelisted read commands; dangerous ops need approval.

LLM Mart · 0 points · 0 views 0 listing impressions 0 install-command copies
Virus-scanned Reviewed automatically before listing.

Full trust report

Download navinspire-ia-navin-navin_skills_sysops-administrator-e9c73a3.zip · 2 KB
Part of navinspire-ia/navin — 182 skills

Install

skills CLI npx skills add https://github.com/Navinspire-ia/navin/tree/main/navin/skills/sysops-administrator
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install navinspire-ia-navin@llmmart
Git git clone https://github.com/Navinspire-ia/navin.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole navinspire-ia/navin collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

SysOps Administrator

Overview

System administration with a strict read-first whitelist. Diagnostic commands run freely; anything that changes system state is announced and, when destructive, approved first.

Detect the OS first (Linux, macOS, or Windows) and use the matching command set below. Never run bash-only syntax in PowerShell or vice versa.

Safe diagnostic whitelist - Linux

uptime && free -h && df -h
systemctl status <service> --no-pager && systemctl --failed
journalctl -u <service> --since "1 hour ago" --no-pager
ps aux --sort=-%mem | head -15
ss -tulpn
ip addr && ip route
dig <host> && ping -c 3 <host> && traceroute <host>
docker ps && docker logs --tail 100 <container> && docker inspect <container>
openssl s_client -connect <host>:443 -servername <host> </dev/null 2>/dev/null | openssl x509 -noout -dates
last -20 && who

Safe diagnostic whitelist - macOS

uptime && vm_stat && df -h
launchctl list | grep -i <service>
log show --last 1h --predicate 'process == "<service>"' --info
ps aux -m | head -15
lsof -iTCP -sTCP:LISTEN -n -P
ifconfig && netstat -rn
dig <host> && ping -c 3 <host> && traceroute <host>
system_profiler SPSoftwareDataType SPHardwareDataType
softwareupdate --list

Safe diagnostic whitelist - Windows (PowerShell)

Get-ComputerInfo | Select-Object OsName, OsUptime, CsTotalPhysicalMemory
Get-Service <service> ; Get-Service | Where-Object Status -eq 'Stopped' | Where-Object StartType -eq 'Automatic'
Get-WinEvent -LogName System -MaxEvents 50 | Where-Object LevelDisplayName -in 'Error','Critical'
Get-Process | Sort-Object WS -Descending | Select-Object -First 15
Get-NetTCPConnection -State Listen | Select-Object LocalAddress, LocalPort, OwningProcess
Get-NetIPAddress ; Get-NetRoute
Resolve-DnsName <host> ; Test-NetConnection <host> -Port 443
Get-PSDrive -PSProvider FileSystem
Get-ChildItem Cert:\LocalMachine\My | Select-Object Subject, NotAfter
Get-HotFix | Sort-Object InstalledOn -Descending | Select-Object -First 10

Workflow

  1. Detect the OS and shell (bash/zsh vs PowerShell), then baseline: load, memory, disk, failed services, recent error logs.
  2. Narrow down: correlate the failing service with its logs, ports, dependencies, and recent package/config changes.
  3. Fix with the smallest action: config reload before restart, restart before reboot.
  4. Verify: the service answers, the symptom is gone, nothing else broke (re-check failed services).
  5. For recurring issues, leave something durable: monitoring, a scheduled check (cron / launchd / Task Scheduler), log rotation, an alert.

Service management per OS

Action Linux macOS Windows
Status systemctl status <svc> launchctl list \| grep <svc> Get-Service <svc>
Logs journalctl -u <svc> log show --predicate 'process == "<svc>"' Get-WinEvent -ProviderName <svc>
Restart* systemctl restart <svc> launchctl kickstart -k <svc> Restart-Service <svc>

*Restart of shared services requires human-approval.

Rules

  • Require human-approval for: rm/Remove-Item outside the workspace, shutdown/reboot/Restart-Computer, firewall changes (iptables/nft/pfctl/Set-NetFirewallRule), volume or partition operations, user/permission changes, package removals, and stopping shared services.
  • Never edit a config file without keeping a .bak copy and stating the rollback command.
  • Certificates: report expiry dates proactively when inspecting TLS services.
  • On Windows prefer PowerShell over cmd.exe; on macOS remember BSD userland flags differ from GNU (e.g. ps aux -m, no free).
Files (navin)
  • SKILL.md 3.9 KB
    ---
    name: sysops-administrator
    description: Administer Linux, macOS, and Windows servers and VMs - services, disks, memory, processes, network, certificates, backups, patches. Whitelisted read commands; dangerous ops need approval.
    metadata: {"navin":{"emoji":"🖥️","category":"devops"}}
    ---
    
    # SysOps Administrator
    
    ## Overview
    
    System administration with a strict read-first whitelist. Diagnostic commands run freely; anything that changes system state is announced and, when destructive, approved first.
    
    **Detect the OS first** (Linux, macOS, or Windows) and use the matching command set below. Never run bash-only syntax in PowerShell or vice versa.
    
    ## Safe diagnostic whitelist - Linux
    
    ```bash
    uptime && free -h && df -h
    systemctl status <service> --no-pager && systemctl --failed
    journalctl -u <service> --since "1 hour ago" --no-pager
    ps aux --sort=-%mem | head -15
    ss -tulpn
    ip addr && ip route
    dig <host> && ping -c 3 <host> && traceroute <host>
    docker ps && docker logs --tail 100 <container> && docker inspect <container>
    openssl s_client -connect <host>:443 -servername <host> </dev/null 2>/dev/null | openssl x509 -noout -dates
    last -20 && who
    ```
    
    ## Safe diagnostic whitelist - macOS
    
    ```bash
    uptime && vm_stat && df -h
    launchctl list | grep -i <service>
    log show --last 1h --predicate 'process == "<service>"' --info
    ps aux -m | head -15
    lsof -iTCP -sTCP:LISTEN -n -P
    ifconfig && netstat -rn
    dig <host> && ping -c 3 <host> && traceroute <host>
    system_profiler SPSoftwareDataType SPHardwareDataType
    softwareupdate --list
    ```
    
    ## Safe diagnostic whitelist - Windows (PowerShell)
    
    ```powershell
    Get-ComputerInfo | Select-Object OsName, OsUptime, CsTotalPhysicalMemory
    Get-Service <service> ; Get-Service | Where-Object Status -eq 'Stopped' | Where-Object StartType -eq 'Automatic'
    Get-WinEvent -LogName System -MaxEvents 50 | Where-Object LevelDisplayName -in 'Error','Critical'
    Get-Process | Sort-Object WS -Descending | Select-Object -First 15
    Get-NetTCPConnection -State Listen | Select-Object LocalAddress, LocalPort, OwningProcess
    Get-NetIPAddress ; Get-NetRoute
    Resolve-DnsName <host> ; Test-NetConnection <host> -Port 443
    Get-PSDrive -PSProvider FileSystem
    Get-ChildItem Cert:\LocalMachine\My | Select-Object Subject, NotAfter
    Get-HotFix | Sort-Object InstalledOn -Descending | Select-Object -First 10
    ```
    
    ## Workflow
    
    1. Detect the OS and shell (bash/zsh vs PowerShell), then baseline: load, memory, disk, failed services, recent error logs.
    2. Narrow down: correlate the failing service with its logs, ports, dependencies, and recent package/config changes.
    3. Fix with the smallest action: config reload before restart, restart before reboot.
    4. Verify: the service answers, the symptom is gone, nothing else broke (re-check failed services).
    5. For recurring issues, leave something durable: monitoring, a scheduled check (cron / launchd / Task Scheduler), log rotation, an alert.
    
    ## Service management per OS
    
    | Action | Linux | macOS | Windows |
    |---|---|---|---|
    | Status | `systemctl status <svc>` | `launchctl list \| grep <svc>` | `Get-Service <svc>` |
    | Logs | `journalctl -u <svc>` | `log show --predicate 'process == "<svc>"'` | `Get-WinEvent -ProviderName <svc>` |
    | Restart* | `systemctl restart <svc>` | `launchctl kickstart -k <svc>` | `Restart-Service <svc>` |
    
    *Restart of shared services requires `human-approval`.
    
    ## Rules
    
    - Require `human-approval` for: `rm`/`Remove-Item` outside the workspace, `shutdown`/`reboot`/`Restart-Computer`, firewall changes (`iptables`/`nft`/`pfctl`/`Set-NetFirewallRule`), volume or partition operations, user/permission changes, package removals, and stopping shared services.
    - Never edit a config file without keeping a `.bak` copy and stating the rollback command.
    - Certificates: report expiry dates proactively when inspecting TLS services.
    - On Windows prefer PowerShell over cmd.exe; on macOS remember BSD userland flags differ from GNU (e.g. `ps aux -m`, no `free`).
    

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.

Related