agentsclimarketplace

Docker setup

Skill viknesh20-20/claude-code-tool-kit/.claude/skills/docker-setup

Production-ready Claude Code configuration. 12 original agents, 200+ slash-command skills, 45+ MCP servers, 14 plugins, design + 3D + WebGPU + GSAP + RAG tooling. One-command Node.js installer for premium websites, SaaS apps, AI agents. Free, MIT, stack-agnostic.

Install
npx -y skills add viknesh20-20/claude-code-tool-kit --skill docker-setup

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Generates Dockerfile, docker-compose.yml, and .dockerignore for the project. Auto-detects the stack and creates optimized multi-stage builds with caching, health checks, and dev/prod configurations.

SKILL.md

2.7 KB, as published. Nobody here has run it

Docker Setup Generator

Detect Stack

!ls package.json requirements.txt pyproject.toml go.mod Cargo.toml *.csproj Gemfile composer.json mix.exs 2>/dev/null !ls Dockerfile docker-compose* .dockerignore 2>/dev/null !cat package.json 2>/dev/null | grep -E '"(main|scripts)"' -A 5 | head -10


Generation Process

Step 1: Analyze Requirements

From project files, determine:

  • Base image (node:20-alpine, python:3.12-slim, golang:1.22, rust:1.75, etc.)
  • Build steps (install deps, compile, bundle)
  • Runtime dependencies (only what's needed to run, not to build)
  • Port(s) the application listens on
  • Required services (database, cache, message queue)
  • Environment variables needed

Step 2: Generate Dockerfile

Multi-stage build pattern:

# Stage 1: Dependencies
FROM base AS deps
COPY lock-file .
RUN install-command

# Stage 2: Build
FROM deps AS build
COPY source .
RUN build-command

# Stage 3: Production
FROM base-slim AS production
COPY --from=build /app/dist ./dist
CMD ["start-command"]

Best practices to include:

  • Use specific image tags, not latest
  • Use .dockerignore to minimize context
  • Copy lock files before source (layer caching)
  • Run as non-root user
  • Set HEALTHCHECK
  • Use multi-stage builds to minimize final image size
  • Set appropriate EXPOSE port
  • Use ENTRYPOINT + CMD pattern

Step 3: Generate docker-compose.yml

services:
  app:
    build: .
    ports: ["3000:3000"]
    environment: ...
    depends_on: ...
    volumes: ...  # for dev: mount source code
    healthcheck: ...

  # Add required services
  db:  # if database detected
  redis:  # if cache detected

Include:

  • Development overrides (volume mounts, hot reload)
  • Named volumes for persistent data
  • Network configuration
  • Health checks for all services
  • Environment variable references (not hardcoded values)

Step 4: Generate .dockerignore

node_modules/
.git/
dist/
build/
*.log
.env
.env.*

Step 5: Document

Add comments explaining:

  • Why each stage exists
  • How to build: docker build -t app .
  • How to run: docker compose up
  • How to access: http://localhost:PORT
  • How to run tests in Docker

Rules

  • Never hardcode secrets in Dockerfile or docker-compose
  • Always use non-root user in production images
  • Always include health checks
  • Minimize image size (alpine/slim base, multi-stage)
  • Include .dockerignore to speed up builds

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.