Ask Claude to Write a Dockerfile by Reading How Your App Actually Runs

Writing a good Dockerfile means understanding your app's build process, runtime dependencies, and production needs. Claude reads all of that from your project.

"Read my project and write a production-ready Dockerfile"

Claude inspects your package files, build scripts, runtime config, and entry points, then generates a Dockerfile that actually fits — not a generic template with placeholder comments.

For a Node.js app, it might produce:

FROM node:22-alpine AS build
WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build

FROM node:22-alpine
WORKDIR /app
RUN addgroup -S app && adduser -S app -G app
COPY --from=build /app/dist ./dist
COPY --from=build /app/node_modules ./node_modules
COPY package*.json ./
USER app
EXPOSE 3000
CMD ["node", "dist/server.js"]

You can be specific about what you need:

# Development Dockerfile with hot reloading
"Write a Dockerfile for local development with volume mounts and live reload"

# Multi-service setup
"Create a docker-compose.yml with my app, a Postgres database, and Redis"

# Optimized for size
"Write the smallest possible Dockerfile for this Go service"

# With health checks
"Add a health check endpoint and HEALTHCHECK instruction to the Dockerfile"

Claude handles the details that are easy to get wrong:

• Multi-stage builds — keeps the final image small by separating build and runtime
• Layer caching — copies dependency files first so installs are cached
• Security — runs as non-root, avoids including secrets or dev dependencies
• .dockerignore — generates one alongside the Dockerfile to exclude node_modules, .git, and test files

If you already have a Dockerfile that's slow or oversized, paste it and ask Claude to optimize — it spots unnecessary layers, missing cache opportunities, and bloated base images.

A good Dockerfile mirrors how your app runs — Claude reads the source of truth and writes one that fits.

via Claude Code