NixOS Commands - Quick Reference¶

Command Cheat Sheet¶

Daily Operations¶

# Start Claude Code
claude

# Check system health
/system-health-check

# Review open tasks
/check_tasks

Weekly Maintenance¶

# Monday: Health check
/system-health-check

# Wednesday: Update flake inputs
/flake-update

# Friday: Review tasks
/check_tasks

Monthly Maintenance¶

# Week 1: Configuration audit
/config-audit

# Week 2: Update Claude Code
/update-claude-code

# Week 3: Update other packages
/update-package

# Week 4: Deploy and verify
/deploy-all

Command Syntax¶

Command	Purpose	Duration	Risk
`/flake-update`	Update flake inputs	10-15m	Med
`/system-health-check`	Infrastructure audit	5-10m	None
`/update-claude-code`	Update Claude Code	8-12m	Low
`/update-package`	Update any package	10-20m	Med
`/deploy-all`	Deploy to all hosts	5-15m	Med-High
`/config-audit`	Quality audit	15-25m	None

Common Workflows¶

Update and Deploy¶

# 1. Health check
/system-health-check

# 2. Update
/flake-update

# 3. Deploy
/deploy-all

# 4. Verify
/system-health-check

Package Update¶

# 1. Update package
/update-package

# 2. Review changes
/review

# 3. Deploy
/deploy-all

Quality Maintenance¶

# 1. Audit
/config-audit

# 2. Create issues
/new_task

# 3. Verify improvements
/config-audit

Just Commands¶

Testing¶

just check-syntax     # Syntax validation
just validate-quick   # Quick validation
just quick-test       # Test all hosts (parallel)
just test-host HOST   # Test specific host

Deployment¶

just quick-deploy HOST      # Smart deploy (if changed)
just deploy-all-parallel    # Deploy all (parallel)
just p620                   # Deploy to p620
just razer                  # Deploy to razer
just p510                   # Deploy to p510

Monitoring¶

grafana-status           # Grafana status
prometheus-status        # Prometheus status
node-exporter-status     # Exporter status

Emergency Procedures¶

Fast Rollback¶

# Rollback single host
ssh HOST "sudo nixos-rebuild switch --rollback"

# Rollback all hosts
for host in p620 razer p510; do
  ssh $host "sudo nixos-rebuild switch --rollback"
done

Check System State¶

# Service failures
ssh HOST "systemctl --failed"

# Recent logs
ssh HOST "journalctl -xe -n 100"

# Current generation
ssh HOST "readlink /nix/var/nix/profiles/system"

Emergency Deployment¶

# Skip tests (use cautiously)
just emergency-deploy HOST

Monitoring Access¶

Web Interfaces¶

# Grafana
http://p620:3001
# Login: admin / nixos-admin

# Prometheus
http://p620:9090

# Alertmanager
http://p620:9093

CLI Checks¶

# Prometheus targets
curl -s http://p620:9090/api/v1/targets | jq

# Active alerts
curl -s http://p620:9093/api/v2/alerts | jq

# Service status
ssh HOST "systemctl status SERVICE"

GitHub Operations¶

Issues and Tasks¶

# Create new task
/new_task

# Check open tasks
/check_tasks

# Review code
/review

Branches and Commits¶

# Branch naming
<type>/<issue-num>-<description>

# Examples:
chore/123-update-claude-code
fix/67-p510-boot-delay
feat/156-new-feature

# Commit format
<type>(<scope>): <description> (#issue)

# Example:
chore(claude-code): update to 2.0.57 (#123)

Pull Requests¶

# Create PR
gh pr create --fill

# Merge PR
gh pr merge NUM --squash --delete-branch

# View PR
gh pr view NUM

File Locations¶

Commands¶

.claude/commands/
├── flake-update.md
├── update-claude-code.md
├── system-health-check.md
├── update-package.md
├── deploy-all.md
└── config-audit.md

Documentation¶

docs/Nixos/
├── README.md
├── Command-System-Overview.md
├── Quick-Reference.md
└── [Command guides]

Important Docs¶

docs/PATTERNS.md              # Best practices
docs/NIXOS-ANTI-PATTERNS.md   # What to avoid
docs/GITHUB-WORKFLOW.md       # GitHub workflow
.agent-os/product/roadmap.md  # Project roadmap

Host Information¶

Host	Role	Hardware	IP
P620	Workstation/Monitoring	AMD Ryzen/ROCm	192.168.1.x
Razer	Mobile/Development	Intel/NVIDIA	192.168.1.x
P510	Media Server	Intel Xeon/NVIDIA	192.168.1.x

Critical Services¶

P620 (Monitoring Server):

Prometheus (9090)
Grafana (3001)
Alertmanager (9093)
AI Services (Ollama)

P510 (Media Server):

Plex
NZBGet
Tautulli

Troubleshooting¶

Build Failures¶

# Check syntax
just check-syntax

# Show detailed error
nix build --show-trace

# Keep failed build
nix build --keep-failed

Service Failures¶

# Check status
systemctl status SERVICE

# View logs
journalctl -u SERVICE -n 100

# Restart service
systemctl restart SERVICE

Deployment Issues¶

# Test configuration
just test-host HOST

# Check differences
just diff HOST

# Rollback
ssh HOST "sudo nixos-rebuild switch --rollback"

Performance Tips¶

Fast Testing¶

# Parallel testing (75% faster)
just quick-test

# Instead of sequential:
just test-all

Smart Deployment¶

# Only deploy if changed
just quick-deploy HOST

# Instead of always deploying:
just HOST

Build Optimization¶

# Use binary caches
# Already configured in flake.nix

# Garbage collection
for host in p620 razer p510; do
  ssh $host "nix-collect-garbage -d"
done

Keyboard Shortcuts¶

Claude Code¶

# In Claude Code session:
Ctrl+C          # Interrupt current operation
Ctrl+D          # Exit Claude Code
/help           # Show help
/clear          # Clear conversation

Tmux (if using)¶

Ctrl+B %        # Split vertically
Ctrl+B "        # Split horizontally
Ctrl+B arrow    # Navigate panes
Ctrl+B d        # Detach session

Safety Checklist¶

Before Deployment¶

Configuration tested: just quick-test
Git status clean: git status
No critical issues: /check_tasks
Monitoring operational: grafana-status
Rollback plan ready

After Deployment¶

Services running: Check systemctl --failed
Monitoring updated: Check Grafana dashboards
No new alerts: Check Alertmanager
Documentation updated
GitHub issues closed

Quick Diagnostics¶

# System health
/system-health-check

# Disk usage
for host in p620 razer p510; do
  ssh $host "df -h / /nix/store"
done

# Service status
for host in p620 razer p510; do
  ssh $host "systemctl --failed"
done

# Recent errors
for host in p620 razer p510; do
  ssh $host "journalctl -p err -n 10 --no-pager"
done

Support Resources¶

Documentation¶

README - Getting started
Overview - Complete guide
Patterns - Best practices
Anti-patterns - What to avoid

Tools¶

Just: Task runner (just --list)
GitHub CLI: gh --help
Claude Code: claude --help

Monitoring¶

Grafana: Visualization and dashboards
Prometheus: Metrics collection
Alertmanager: Alert management

Remember: When in doubt, check /system-health-check first!