Agent Skills: Docker Best Practices Skill

Docker Best Practices Skill

When to Use

Use this skill when:

• Creating new Dockerfiles
• Optimizing existing container images
• Setting up docker-compose environments
• Implementing health checks
• Securing container deployments
• Reducing image sizes
• Configuring multi-stage builds

Dockerfile Patterns

1. Multi-Stage Build (Python)

# Stage 1: Builder
FROM python:3.12-slim AS builder

WORKDIR /app

# Install build dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential \
    && rm -rf /var/lib/apt/lists/*

# Install Python dependencies
COPY requirements.txt .
RUN pip wheel --no-cache-dir --wheel-dir /wheels -r requirements.txt

# Stage 2: Runtime
FROM python:3.12-slim AS runtime

WORKDIR /app

# Create non-root user
RUN groupadd -r appgroup && useradd -r -g appgroup appuser

# Copy wheels from builder
COPY --from=builder /wheels /wheels
RUN pip install --no-cache-dir /wheels/* && rm -rf /wheels

# Copy application
COPY --chown=appuser:appgroup . .

# Switch to non-root user
USER appuser

# Health check
HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
    CMD curl -f http://localhost:8000/health || exit 1

EXPOSE 8000

CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]

2. Layer Caching Optimization

# ✅ Good: Dependencies first (cached if unchanged)
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY . .

# ❌ Bad: Entire context first (cache invalidated on any change)
COPY . .
RUN pip install -r requirements.txt

3. Security Hardening

# ✅ Security checklist:
# 1. Non-root user
USER appuser

# 2. Read-only filesystem
# docker run --read-only ...

# 3. No new privileges
# docker run --security-opt=no-new-privileges ...

# 4. Drop capabilities
# docker run --cap-drop=ALL ...

# 5. Minimal base image
FROM python:3.12-alpine  # or distroless

# 6. No secrets in image
# Use environment variables or secrets mount

4. Health Checks

# HTTP health check
HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
    CMD curl -f http://localhost:8000/health || exit 1

# TCP health check (no curl)
HEALTHCHECK --interval=30s --timeout=10s --retries=3 \
    CMD nc -z localhost 8000 || exit 1

# Script health check
HEALTHCHECK CMD ["/app/healthcheck.sh"]

Docker Compose Patterns

Development Environment

version: "3.9"

services:
  app:
    build:
      context: .
      dockerfile: Dockerfile.dev
    volumes:
      - .:/app:cached
      - /app/__pycache__  # Exclude pycache
    ports:
      - "8000:8000"
    environment:
      - DEBUG=true
      - DATABASE_URL=postgresql://...
    depends_on:
      db:
        condition: service_healthy
    
  db:
    image: postgres:16-alpine
    volumes:
      - postgres_data:/var/lib/postgresql/data
    environment:
      - POSTGRES_DB=app
      - POSTGRES_USER=app
      - POSTGRES_PASSWORD=secret
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U app"]
      interval: 5s
      timeout: 5s
      retries: 5

volumes:
  postgres_data:

Production Environment

version: "3.9"

services:
  app:
    image: myapp:${VERSION:-latest}
    deploy:
      replicas: 3
      resources:
        limits:
          cpus: '0.5'
          memory: 512M
      restart_policy:
        condition: on-failure
        max_attempts: 3
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
      interval: 30s
      timeout: 10s
      retries: 3
    networks:
      - app-network
    logging:
      driver: "json-file"
      options:
        max-size: "10m"
        max-file: "3"

Image Size Optimization

✅ Size reduction techniques:
1. Use slim/alpine base images
2. Multi-stage builds
3. Combine RUN commands
4. Clean up apt cache: && rm -rf /var/lib/apt/lists/*
5. Use .dockerignore
6. Remove build dependencies after install
7. Use --no-cache-dir for pip

Output Format

## Docker Analysis Report

### Image Analysis
- Base image: python:3.12-slim
- Final size: 125MB
- Layers: 12
- Security: ✅ Non-root user

### Optimization Suggestions
1. [Current] → [Optimized] - [Size saved]

### Security Findings
- ✅ Non-root user configured
- ⚠️ Health check missing
- ❌ Running as root

### docker-compose Review
- ✅ Health checks defined
- ✅ Resource limits set
- ⚠️ No restart policy

Example Usage

@docker Create optimized Dockerfile for FastAPI app
@docker Review docker-compose.yml for production readiness
@docker Reduce image size of the current Dockerfile
@docker Add health checks to all services

GPU Inference Container Patterns

LLM Serving Container (vLLM)

FROM nvidia/cuda:12.4.0-base-ubuntu22.04

RUN apt-get update && apt-get install -y --no-install-recommends \
    python3.11 python3-pip curl \
    && rm -rf /var/lib/apt/lists/*

# Install vLLM for GPU inference
RUN pip install vllm==0.6.0

# Create non-root user
RUN useradd -m -u 1000 -s /bin/bash inference
USER inference
WORKDIR /app

COPY --chown=inference:inference .

# Expose inference API port
EXPOSE 8000

# Health check: verify model loaded and API responsive
HEALTHCHECK --interval=30s --timeout=10s --start-period=120s --retries=3 \
    CMD curl -f http://localhost:8000/health || exit 1

# Start vLLM server with GPU
ENTRYPOINT ["python3", "-m", "vllm.entrypoints.openai.api_server"]
CMD ["--model", "mistralai/Mistral-7B-Instruct-v0.3", "--host", "0.0.0.0"]

Ollama Development Container

services:
  ollama:
    image: ollama/ollama:latest
    container_name: ollama-dev
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    volumes:
      - ollama_data:/root/.ollama
      - ./models:/models  # Custom model mounts
    ports:
      - "11434:11434"
    healthcheck:
      test: ["CMD", "ollama", "list"]
      interval: 30s
      timeout: 10s
      retries: 3
    environment:
      - OLLAMA_HOST=0.0.0.0
      - OLLAMA_KEEP_ALIVE=5m

volumes:
  ollama_data:

Agent Sandboxing Pattern

Isolated Agent Execution Container

FROM python:3.12-slim

# Security hardening
RUN useradd -m -u 1000 agent
RUN chmod 700 /home/agent

# No root, no sudo, no unnecessary packages
RUN apt-get update && apt-get install -y --no-install-recommends \
    ca-certificates \
    && rm -rf /var/lib/apt/lists/* \
    && apt-get remove -y --purge sudo

# Read-only filesystem compatible
WORKDIR /home/agent/app

# Drop all capabilities
RUN setcap -r /usr/bin/python3.12

USER agent
COPY --chmod=500 --chown=agent:agent .

# Agent sandbox constraints
ENV PYTHONPATH=/home/agent/app
ENV PYTHONWRITEBYTECODE=1
ENV MALLOC_CHECK_=2

HEALTHCHECK --interval=30s --timeout=5s --start-period=10s \
    CMD python3 -c "import sys; sys.exit(0)" || exit 1

ENTRYPOINT ["python3"]
CMD ["agent_entrypoint.py"]