deploy-planner

Deployment and DevOps agent that generates Dockerfiles, CI/CD configs, and step-by-step deployment guides for free hosting platforms. Triggers on: deploy, launch, hosting, Docker, CI/CD, production, g

SKILL.md

GitHub

Deploy Planner — Launch Preparation Agent

Purpose

Prepare the complete deployment package so the user can go from "code on my machine" to "live on the internet" with clear, step-by-step instructions.

When to Use

  • After security audit is complete and approved
  • When the user asks "how do I deploy this?"
  • As the final phase of the SaaS Accelerator pipeline

Process

Step 1: Deployment Target Selection

Recommend a free-tier hosting platform based on the tech stack:

Tech Stack Recommended Host Free Tier Limits
Next.js Vercel 100GB bandwidth/month
Node.js/Express Render 750 hours/month
Python/Django Railway $5 free credit/month
Static + API Netlify + Supabase 100GB + 500MB DB

Step 2: Docker Configuration

Generate a production Dockerfile:

  • Multi-stage build (build stage + production stage)
  • Non-root user
  • Health check endpoint
  • Minimal image size

Generate a docker-compose.yml for local development:

  • App service
  • Database service
  • Volume mounts for persistence
  • Environment variable passthrough

Step 3: Environment Setup Guide

Create a .env.production.example with:

  • Every required variable
  • Description of what each variable does
  • Where to obtain each value (e.g., "Get from Supabase dashboard")

Step 4: Deployment Guide

Write a step-by-step guide:

## Deploy to [Platform]

### Prerequisites
- [ ] Account on [platform]
- [ ] Git repository pushed to GitHub

### Steps
1. Go to [platform dashboard URL]
2. Click "New Project" → "Import from GitHub"
3. Select your repository
4. Set environment variables: [list each one]
5. Click "Deploy"
6. Wait for build to complete (~2-3 minutes)
7. Visit your live URL: https://your-app.vercel.app

### Post-Deployment Verification
- [ ] Homepage loads correctly
- [ ] User registration works
- [ ] Login/logout works
- [ ] Core feature works end-to-end
- [ ] No console errors in browser DevTools

Step 5: Project README

Generate the final README.md for the project repository:

  • Project name and one-line description
  • Tech stack badges
  • Features list
  • Local development setup (step-by-step)
  • Deployment instructions
  • Environment variables reference
  • Contributing guidelines
  • License

Output

  • Dockerfile
  • docker-compose.yml
  • .env.production.example
  • deploy_guide.md
  • Updated README.md

Exit Criteria

  • Deployment guide is platform-specific and step-by-step
  • Docker configuration is production-ready
  • All environment variables are documented
  • Post-deployment verification checklist exists
  • README is complete and professional