infra-ops

Infrastructure operations skills for Claude Code. Turn your coding agent into a cross-platform sysadmin that diagnoses, triages, and resolves system issues systematically.

What it does

Instead of guessing or running random commands, infra-ops gives Claude Code structured methodologies for infrastructure work:

Core Diagnostics:

• system-diagnostics - Comprehensive health check with severity ratings
• incident-triage - Structured hypothesis-driven troubleshooting
• performance-optimizer - Profile, identify bottlenecks, optimize with before/after metrics
• log-analyzer - Parse and correlate logs across sources and time windows
• network-troubleshooter - OSI-layer systematic network diagnosis

Infrastructure Tools:

• container-ops - Docker and Kubernetes container troubleshooting
• security-auditor - System security posture audit with scoring
• service-monitor - Service health checks and endpoint monitoring
• disk-cleanup - Safe disk space recovery with before/after metrics
• ssl-checker - SSL/TLS certificate diagnostics and chain verification

Key features

• Cross-platform - Windows (PowerShell), Linux, macOS
• Runs real commands - Actual diagnostics, not just suggestions
• Safe by default - Read-only checks; asks permission before any changes
• Structured reports - Consistent, formatted output for every skill
• Escalation-aware - Knows when to ask for help or more access
• 10 skills - Complete coverage from network to containers to security

Installation

Claude Code (via Plugin)

/plugin marketplace add skyvanguard/infra-ops
/plugin install infra-ops@infra-ops

Manual Installation

Clone into your Claude Code plugins directory:

cd ~/.claude/plugins
git clone https://github.com/skyvanguard/infra-ops.git

Usage

Skills trigger automatically based on context, or invoke them directly:

# System health check
"check my system health"
"run a diagnostic"

# Incident triage
"my database isn't responding"
"nginx is returning 502 errors"

# Performance optimization
"my computer is slow, optimize it"
"what's eating all my RAM?"

# Log analysis
"check the logs for errors in the last hour"
"what happened at 3am yesterday?"

# Network troubleshooting
"I can't reach api.example.com"
"my internet is intermittent"

# Container operations
"why is my docker container restarting?"
"check kubernetes pod status"

# Security audit
"run a security check on this system"
"are there any exposed ports?"

# Service monitoring
"check if all services are healthy"
"is the API responding?"

# Disk cleanup
"my disk is almost full, help me free space"
"what's using all my storage?"

# SSL/TLS checks
"check the SSL certificate for example.com"
"when does our cert expire?"

Skills in detail

system-diagnostics

Checks CPU, memory, disk, network interfaces, top processes, and uptime. Rates each category (OK/WARNING/CRITICAL) and produces a formatted report with recommendations.

incident-triage

Follows a hypothesis-driven process: classify -> gather context -> form hypotheses -> test -> identify root cause -> present resolution options. Includes escalation criteria.

performance-optimizer

Profiles current resource usage, identifies bottlenecks by threshold, presents a prioritized optimization plan (quick wins first), executes approved changes, and verifies improvement with before/after metrics.

log-analyzer

Identifies relevant log sources, filters by time/severity/keywords, analyzes patterns (burst, gradual, periodic), correlates events across sources, and builds an event timeline.

network-troubleshooter

Follows OSI layers bottom-up: physical/link -> network -> transport -> application. Verifies each layer before moving up. Identifies the exact failure point with evidence.

container-ops

Diagnoses Docker and Kubernetes issues: container health, resource usage, networking, image problems, and common failure patterns (CrashLoopBackOff, OOM kills, ImagePullBackOff).

security-auditor

Audits system security: open ports, user privileges, SSH config, pending updates, file permissions, and running services. Produces a scored security report with prioritized remediation.

service-monitor

Checks service health across HTTP endpoints, TCP ports, databases, and processes. Measures response times, maps dependencies, and produces a service health dashboard.

disk-cleanup

Identifies disk space consumers (temp files, caches, containers, logs, old downloads) and safely reclaims storage. Shows before/after comparison and never deletes without permission.

ssl-checker

Verifies SSL/TLS certificates: expiration, chain validity, protocol versions, cipher strength, and SAN coverage. Diagnoses common browser SSL errors.

Safety

All skills follow these principles:

1. Read-only by default - Diagnostic commands only gather information
2. Explicit permission - Any system modification requires user approval
3. Show before execute - Commands are displayed before running
4. Never destructive - Won't delete data, disable security, or modify critical configs
5. Transparent - All commands and their purpose are explained

Contributing

1. Fork the repository
2. Create a branch for your skill or improvement
3. Submit a PR

Ideas for future skills:

• backup-validator - Verify backup integrity and recovery procedures
• cloud-cost-analyzer - AWS/GCP/Azure resource cost analysis
• dns-manager - DNS record troubleshooting and validation
• cron-debugger - Scheduled task troubleshooting

License

MIT License - see LICENSE file for details.