containerizing-applications
About
This Claude skill helps developers containerize applications using Docker, Docker Compose, and Helm charts. It provides best practices like hardened base images, multi-stage builds, and covers common pitfalls. Use it when creating or optimizing container configurations for deployment.
Quick Install
Claude Code
Recommended/plugin add https://github.com/majiayu000/claude-skill-registrygit clone https://github.com/majiayu000/claude-skill-registry.git ~/.claude/skills/containerizing-applicationsCopy and paste this command in Claude Code to install this skill
Documentation
Containerizing Applications
Quick Start
- Run impact analysis first (env vars, network topology, auth/CORS)
- Generate Dockerfiles using patterns below
- Create docker-compose.yml with proper networking
- Package as Helm chart for Kubernetes
Dockerfile Patterns
FastAPI/Python (Multi-stage with uv)
# syntax=docker/dockerfile:1
FROM python:3.13-slim AS builder
WORKDIR /app
RUN pip install uv
COPY pyproject.toml .
RUN uv pip install --system --no-cache -r pyproject.toml
FROM python:3.13-slim
WORKDIR /app
COPY --from=builder /usr/local/lib/python3.13/site-packages /usr/local/lib/python3.13/site-packages
COPY . .
RUN useradd -u 1000 appuser && chown -R appuser /app
USER appuser
EXPOSE 8000
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]
Next.js (Standalone)
# syntax=docker/dockerfile:1
FROM node:20-alpine AS deps
WORKDIR /app
COPY package*.json ./
RUN npm ci
FROM node:20-alpine AS builder
WORKDIR /app
COPY --from=deps /app/node_modules ./node_modules
COPY . .
ARG NEXT_PUBLIC_API_URL
ARG NEXT_PUBLIC_SSO_URL
ENV NEXT_PUBLIC_API_URL=$NEXT_PUBLIC_API_URL
ENV NEXT_PUBLIC_SSO_URL=$NEXT_PUBLIC_SSO_URL
RUN npm run build
FROM node:20-alpine AS runner
WORKDIR /app
ENV NODE_ENV=production
RUN addgroup -g 1001 -S nodejs && adduser -S nextjs -u 1001
COPY --from=builder /app/.next/standalone ./
COPY --from=builder /app/.next/static ./.next/static
COPY --from=builder /app/public ./public
USER nextjs
EXPOSE 3000
CMD ["node", "server.js"]
docker-compose Pattern
services:
web:
build:
context: ./web
args:
# BROWSER: baked into JS bundle
- NEXT_PUBLIC_API_URL=http://localhost:8000
environment:
# SERVER: read at runtime inside container
- SERVER_API_URL=http://api:8000
ports:
- "3000:3000"
depends_on:
api:
condition: service_healthy
api:
build: ./api
environment:
- DATABASE_URL=${DATABASE_URL}
- CORS_ORIGINS=http://localhost:3000,http://web:3000
ports:
- "8000:8000"
healthcheck:
test: ["CMD", "curl", "-f", "http://127.0.0.1:8000/health"]
interval: 30s
timeout: 10s
retries: 3
start_period: 40s
Helm Chart Structure
helm/myapp/
├── Chart.yaml
├── values.yaml
├── templates/
│ ├── _helpers.tpl
│ ├── deployment.yaml
│ ├── service.yaml
│ ├── configmap.yaml
│ ├── secret.yaml
│ └── ingress.yaml
Chart.yaml
apiVersion: v2
name: myapp
version: 1.0.0
appVersion: "1.0.0"
values.yaml Pattern
api:
replicaCount: 1
image:
repository: myapp/api
tag: latest
pullPolicy: IfNotPresent
service:
type: ClusterIP
port: 8000
resources:
limits:
cpu: 500m
memory: 512Mi
Helm Commands
helm create mychart # Create new chart
helm template . --debug # Render templates
helm install myapp ./chart # Install
helm upgrade myapp ./chart # Upgrade
helm install myapp ./chart \
--set api.image.tag=v2.0.0 # Override values
Battle-Tested Gotchas (15+)
1. Browser vs Server URLs
Problem: Browser runs on host, server runs in container
build:
args:
- NEXT_PUBLIC_API_URL=http://localhost:8000 # Browser
environment:
- SERVER_API_URL=http://api:8000 # Server
2. Healthcheck IPv6 Issue
Problem: wget http://localhost:3000 fails with IPv6
healthcheck:
test: ["CMD", "wget", "--spider", "http://127.0.0.1:3000/"] # NOT localhost!
3. MCP Server 421 Misdirected Request
Problem: FastMCP rejects Docker service names
transport_security = TransportSecuritySettings(
allowed_hosts=["127.0.0.1:*", "localhost:*", "mcp-server:*", "0.0.0.0:*"]
)
4. SQLModel Tables Not Created
Problem: Models not imported before create_all()
# MUST import before create_all()
from .models import User, Task, Project # noqa: F401
SQLModel.metadata.create_all(engine)
5. Database Migration Order
Problem: Drizzle db:push drops tables not in schema
Solution: Start postgres → Run Drizzle → Then start API
6. uv Network Timeout
RUN UV_HTTP_TIMEOUT=120 uv pip install --system --no-cache -r pyproject.toml
7. Missing Syntax Directive
# syntax=docker/dockerfile:1 # ALWAYS first line
FROM python:3.13-slim
8. localhost in Container
Use Docker service names (api, web, sso) for server-side, NOT localhost
9. Auth Origins
Add Docker service names to trustedOrigins BEFORE building
10. Service Startup Order
Use depends_on with condition: service_healthy
11. Health Check Timing
Use start_period (e.g., 40s) for apps that take time to start
12. pgAdmin Email Validation
Use valid email like [email protected], not .local domains
13. Playwright in Dependencies
Keep test tools in devDependencies (300MB+ bloat)
14. MCP Health Check 406
Add separate /health endpoint via ASGI middleware
15. Helm Comma Parsing
Use values file instead of --set for comma-containing values
Production Security
Docker Hardened Images (Recommended)
95% fewer CVEs than community images. Free under Apache 2.0.
# BEFORE: Community image with unknown CVEs
FROM python:3.12-slim
# AFTER: Docker Hardened Image
FROM docker.io/docker/python:3.12-dhi
Five Pillars of DHI:
| Pillar | What You Get |
|---|---|
| Minimal Attack Surface | 98% CVE reduction |
| 100% Complete SBOM | SPDX/CycloneDX format |
| SLSA Build Level 3 | Verified provenance |
| OpenVEX | Machine-readable vuln status |
| Cosign Signatures | Cryptographic verification |
Verify signatures:
cosign verify docker.io/docker/python:3.12-dhi
Read SBOM:
docker sbom docker.io/docker/python:3.12-dhi
Trivy Scanning (CI/CD)
- name: Scan for vulnerabilities
run: trivy image --severity HIGH,CRITICAL --exit-code 1 ${{ env.IMAGE }}
Distroless Images (Alternative)
# Python - use gcr.io/distroless/python3-debian12
FROM gcr.io/distroless/python3-debian12
# No shell, no package manager, runs as nonroot by default
Multi-Arch Builds
- uses: docker/build-push-action@v5
with:
platforms: linux/amd64,linux/arm64 # Build for both
cache-from: type=gha
cache-to: type=gha,mode=max
BuildKit Secrets
# Mount secrets during build (never stored in layers)
RUN --mount=type=secret,id=npm_token \
NPM_TOKEN=$(cat /run/secrets/npm_token) npm install
See references/production-security.md for full patterns.
Verification
Run: python scripts/verify.py
Related Skills
operating-k8s-local- Local Kubernetes with Minikubedeploying-cloud-k8s- Cloud Kubernetes deploymentscaffolding-fastapi-dapr- FastAPI patterns
References
- references/production-security.md - Trivy, distroless, multi-arch, BuildKit secrets
GitHub Repository
Related Skills
content-collections
MetaThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
creating-opencode-plugins
MetaThis skill provides the structure and API specifications for creating OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It offers implementation patterns for JavaScript/TypeScript modules that intercept and extend the AI assistant's lifecycle. Use it when you need to build event-driven plugins for monitoring, custom handling, or extending OpenCode's capabilities.
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
