docker-ops

name	docker-ops
description	Standardized container management and orchestration using NAVIG's docker primitives.
metadata	{"navig":{"emoji":"🐳","requires":{"bins":["navig"]}}}

Docker Operations Skill

Manage containers, logs, and compose stacks on remote hosts via NAVIG.

Container Management

List & Inspect

# List running containers
navig docker ps

# List all containers (including stopped)
navig docker ps -a

# Inspect specific container details
navig docker inspect <container_name_or_id>

Lifecycle Control

# Restart a container
navig docker restart <container_name>

# Stop a container
navig docker stop <container_name>

# Start a container
navig docker start <container_name>

Logs & Debugging

View Logs

Crucial: Always use -n to limit log output.

# Get last 100 lines of logs
navig docker logs <container_name> -n 100

# Get logs since 10 minutes ago
navig docker logs <container_name> --since 10m

Execute Commands

# Run a command inside a container
navig docker exec <container_name> "ls -la /app"

# Run an interactive shell (if supported by agent interface)
navig docker exec <container_name> "sh"

Docker Compose

Stack Management

# Deploy/Update stack (Detached mode is default recommendation)
navig docker compose up -d -f docker-compose.yml

# Stop stack
navig docker compose down

# View stack status
navig docker compose ps

Templates

Standard Web Stack (docker-compose.yml)

version: '3.8'
services:
  app:
    image: myapp:latest
    restart: always
    ports:
      - "3000:3000"
    environment:
      - NODE_ENV=production
  db:
    image: postgres:14-alpine
    volumes:
      - db_data:/var/lib/postgresql/data
volumes:
  db_data:

Best Practices

1. Identify Containers: Use navig docker ps first to get the exact container name.
2. Resource Check: Run navig docker stats --no-stream to check CPU/Memory usage if a container is unresponsive.
3. Pruning: Periodically run navig docker system prune -f (via navig run) to clear space, but ask user first.