Dockerizer

AI-powered Docker configuration generator that creates production-ready Dockerfiles, docker-compose.yml, and .dockerignore files for any repository.

https://dockerizer.dev by Dublyo

Features

• Automatic Stack Detection - Detects language, framework, and version with 90%+ confidence
• Production-Ready Output - Multi-stage builds, non-root users, health checks, optimized layers
• AI Fallback - Uses OpenAI, Anthropic, or Ollama when detection confidence is low
• Interactive Setup - Guided CLI wizard for AI configuration and customization
• Build Plan - Nixpacks-inspired plan command for debugging and transparency
• Procfile Support - Respects Heroku-style Procfiles for start commands
• 26 Providers - Node.js, Python, Go, Rust, Ruby, PHP, Java, .NET, Elixir frameworks supported
• Agent Mode - Iterative analyze → generate → build → test → fix workflow
• MCP Server - Integration with Claude Code and Goose AI assistants
• Recipe System - YAML-based automation workflows

Installation

Quick Install (Recommended)

curl -fsSL https://dockerizer.dev/install.sh | sh

Homebrew (macOS/Linux)

brew install dublyo/tap/dockerizer

Using Go

go install github.com/dublyo/dockerizer/cmd/dockerizer@latest

From Source

git clone https://github.com/dublyo/dockerizer
cd dockerizer/src
make build
sudo make install

Docker

docker run --rm -v $(pwd):/app ghcr.io/dublyo/dockerizer /app

Quick Start

# Interactive setup (recommended for first-time users)
dockerizer init

# Auto-detect and generate Docker configs
dockerizer .

# Preview build plan without generating files
dockerizer plan ./my-project

# Force AI generation for better results
ANTHROPIC_API_KEY=sk-ant-xxx dockerizer --ai ./my-project

Supported Frameworks

Language	Frameworks	Confidence
Node.js	Next.js, NestJS, Nuxt, Remix, Astro, SvelteKit, Hono, Koa, Fastify, Express	80-100%
Python	Django, FastAPI, Flask	90-100%
Go	Gin, Fiber, Echo, Standard	90%
Rust	Actix Web, Axum	90%
Ruby	Rails	85-90%
PHP	Laravel, Symfony	85-95%
Java	Spring Boot, Quarkus	90-95%
.NET	ASP.NET Core	70-90%
Elixir	Phoenix	80-90%

Commands

dockerizer init (Interactive Setup)

Guided wizard that walks you through AI configuration and file generation.

dockerizer init ./my-project

Features:

• Auto-detects your stack and confirms
• Prompts for AI provider (Anthropic, OpenAI, Ollama)
• Securely accepts API keys
• Previews and generates files
• Saves configuration for future use

dockerizer plan (Build Plan)

Output the resolved build plan as JSON or YAML without generating files. Inspired by Nixpacks.

dockerizer plan ./my-project
dockerizer plan --format yaml ./my-project
dockerizer plan --output plan.json ./my-project

The plan includes:

• Detection results (language, framework, version, confidence)
• Build phases with commands
• Cache directories for faster builds
• Start command resolution

dockerizer [path]

Generate Docker configuration files.

dockerizer ./my-project

Flags:

Flag	Description
--ai	Force AI generation even for high-confidence detections
-f, --force	Overwrite existing files
-o, --output	Output directory (default: same as input)
--no-compose	Skip docker-compose.yml generation
--no-ignore	Skip .dockerignore generation
--no-env	Skip .env.example generation
--json	Output results as JSON
-v, --verbose	Enable verbose output
-q, --quiet	Suppress non-essential output

dockerizer detect [path]

Detect stack without generating files.

dockerizer detect ./my-project
dockerizer detect --all ./my-project  # Show all candidates

dockerizer agent [path]

Run in agent mode with iterative build/test/fix cycle.

OPENAI_API_KEY=sk-xxx dockerizer agent ./my-project

dockerizer serve

Start MCP server for AI assistant integration (stdio mode).

dockerizer serve

Configure in Claude Code (~/.claude.json):

{
  "mcpServers": {
    "dockerizer": {
      "command": "dockerizer",
      "args": ["serve"]
    }
  }
}

dockerizer recipe [file]

Execute a YAML workflow recipe.

dockerizer recipe analyze --path ./my-project
dockerizer recipe generate --path ./my-project
dockerizer recipe build-and-test --path ./my-project

dockerizer validate [dockerfile]

Validate Dockerfile syntax and best practices.

dockerizer validate ./Dockerfile

Environment Overrides

Customize build behavior via environment variables (Nixpacks-inspired):

Variable	Description
DOCKERIZER_BUILD_CMD	Override build command
DOCKERIZER_INSTALL_CMD	Override install/setup command
DOCKERIZER_START_CMD	Override start command
DOCKERIZER_APT_PKGS	Additional APT packages (comma-separated)

Example:

DOCKERIZER_START_CMD="gunicorn app:app" dockerizer ./my-project

Output Files

Running dockerizer ./my-project generates:

File	Description
Dockerfile	Multi-stage, optimized, production-ready
docker-compose.yml	Service definition with health checks, resource limits
.dockerignore	Language-specific exclusions
.env.example	Environment variables template

AI Configuration

Configure AI providers via environment variables:

Anthropic (Recommended)

export ANTHROPIC_API_KEY=sk-ant-xxx
export ANTHROPIC_MODEL=claude-3-5-haiku-20241022  # optional

OpenAI

export OPENAI_API_KEY=sk-xxx
export OPENAI_MODEL=gpt-4o-mini  # optional

Ollama (Local)

export OLLAMA_BASE_URL=http://localhost:11434  # optional
export OLLAMA_MODEL=llama3  # optional

AI is automatically used when:

• Detection confidence is below 80%
• --ai flag is specified
• No matching template exists for the detected stack

Configuration File

Create .dockerizer.yml in your project or ~/.dockerizer.yml globally:

ai:
  provider: anthropic
  model: claude-3-5-haiku-20241022

defaults:
  include_compose: true
  include_ignore: true
  include_env: true
  overwrite: false

Example Output

Build Plan (JSON)

{
  "version": "1.0",
  "generator": "dockerizer v1.0.0",
  "detection": {
    "detected": true,
    "language": "nodejs",
    "framework": "nextjs",
    "version": "14.0.0",
    "confidence": 95,
    "provider": "nextjs"
  },
  "phases": [
    {
      "name": "setup",
      "commands": ["npm ci"]
    },
    {
      "name": "build",
      "depends_on": ["setup"],
      "commands": ["npm run build"]
    }
  ],
  "cache_dirs": [
    {"path": "/root/.npm", "id": "npm-cache"}
  ],
  "start": {
    "cmd": "node server.js"
  }
}

Generated Dockerfile

# Build stage
FROM node:20-alpine AS builder
WORKDIR /app
COPY package.json package-lock.json ./
RUN npm ci
COPY . .
ENV NEXT_TELEMETRY_DISABLED=1
RUN npm run build

# Production stage
FROM node:20-alpine AS runner
WORKDIR /app
ENV NODE_ENV=production
RUN addgroup --system --gid 1001 nodejs
RUN adduser --system --uid 1001 nextjs
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"]
HEALTHCHECK --interval=30s --timeout=10s --start-period=40s --retries=3 \
  CMD wget --no-verbose --tries=1 --spider http://localhost:3000/api/health || exit 1

Comparison with Nixpacks

Feature	Dockerizer	Nixpacks
Output	Human-editable Dockerfile	OCI image only
AI Fallback	Yes (OpenAI, Anthropic, Ollama)	No
Interactive Setup	Yes (dockerizer init)	No
Build Plan	Yes (dockerizer plan)	Yes
Agent Mode	Yes (iterative fix)	No
MCP Integration	Yes	No
Procfile Support	Yes	Yes
Cache Directories	Yes	Yes
Env Overrides	Yes (DOCKERIZER_*)	Yes (NIXPACKS_*)

Development

Building

make build          # Build binary to ./build/dockerizer
make build-all      # Build for all platforms
make install        # Install to /usr/local/bin
make clean          # Remove build artifacts

Adding a New Provider

1. Create provider file: providers/<language>/<framework>.go
2. Implement the providers.Provider interface
3. Register in providers/<language>/register.go
4. Add template in internal/generator/generator.go

Roadmap

• More Node.js frameworks (Nuxt, NestJS, Remix, Astro, SvelteKit, Hono, Koa)
• Ruby/Rails support
• PHP/Laravel + Symfony support
• Java/Spring Boot support
• .NET/ASP.NET Core support
• Elixir/Phoenix support
• Kubernetes manifests generation
• GitHub Actions integration
• VS Code extension
• npx-style execution

License

MIT License - see LICENSE for details.

Contributing

Contributions are welcome! Please open an issue or submit a pull request.

Links

• Dockerizer.dev
• Dublyo PaaS
• GitHub
• Issue Tracker