Menu
สร้าง Infrastructure Specification Document สำหรับ DevOps handoff — Docker, server requirements, scaling plan, cost estimation
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
| name | Infrastructure Spec |
| description | สร้าง Infrastructure Specification Document สำหรับ DevOps handoff — Docker, server requirements, scaling plan, cost estimation |
Purpose: สร้าง "Handoff Document สำหรับ DevOps" เหมือนที่
pmo-dev-handoffสร้างสำหรับ Dev ครอบคลุม server spec, Docker config, scaling strategy, estimated cost
สร้างไฟล์ {ProjectFolder}/Scaffold/INFRA-SPEC.md:
┌─────────────┐ ┌──────────────┐ ┌───────────────┐
│ Client │────>│ Load Balancer│────>│ App Server(s)│
│ (Browser/ │ │ (Nginx/ALB) │ │ (Docker) │
│ Mobile) │ └──────────────┘ └───────┬───────┘
└─────────────┘ │
┌───────┴───────┐
│ Database │
│ (PostgreSQL) │
└───────────────┘
| Component | Minimum | Recommended | Notes |
|---|---|---|---|
| App Server | 2 vCPU, 4GB RAM | 4 vCPU, 8GB RAM | Per instance |
| Database | 2 vCPU, 4GB RAM, 50GB SSD | 4 vCPU, 8GB RAM, 100GB SSD | Managed preferred |
| Cache | 1GB Redis | 2GB Redis | Optional for MVP |
| Storage | 20GB | 50GB | File uploads |
# docker-compose.production.yml
version: '3.8'
services:
app:
build:
context: .
dockerfile: Dockerfile
ports:
- "3000:3000"
environment:
- DATABASE_URL=${DATABASE_URL}
- REDIS_URL=${REDIS_URL}
deploy:
replicas: 2
resources:
limits:
cpus: '1.0'
memory: 1G
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:3000/health"]
interval: 30s
timeout: 10s
retries: 3
db:
image: postgres:16
volumes:
- pgdata:/var/lib/postgresql/data
environment:
- POSTGRES_PASSWORD=${DB_PASSWORD}
redis:
image: redis:7-alpine
command: redis-server --maxmemory 256mb
volumes:
pgdata:
| Traffic Level | Strategy | Infra |
|---|---|---|
| MVP (< 100 users) | Single server | 1 app + 1 DB |
| Growth (100-1K users) | Horizontal scale | 2 app + 1 DB + Redis |
| Scale (1K-10K users) | Load balanced | 3+ app + DB read replica + Redis cluster |
| Enterprise (10K+ users) | Auto-scaling | K8s cluster + managed DB + CDN |
| Provider | Service | Estimated Monthly Cost (MVP) |
|---|---|---|
| AWS | ECS Fargate + RDS | $50-150 |
| GCP | Cloud Run + Cloud SQL | $40-120 |
| Vercel + Supabase | Serverless | $25-80 |
| Railway | Managed containers | $20-60 |
| DigitalOcean | Droplet + Managed DB | $30-80 |
| Item | Strategy | Frequency |
|---|---|---|
| Database | Automated snapshot | Daily + before migration |
| File Storage | Cross-region replication | Real-time |
| Application Config | Git versioned | Every change |
| Recovery RTO | < 1 hour | - |
| Recovery RPO | < 24 hours | - |
{ProjectFolder}/Scaffold/INFRA-SPEC.md (AI-Managed)
Pre/Post execution validation + Phase Gate enforcement — ป้องกันการข้าม step, auto-validate output, ให้คะแนน quality score
สแกน code ใน Scaffold/ ด้วย 53 security rules (6 หมวด) — secrets, permissions, injection, AI-specific, config, dependencies — ก่อน deploy หรือหลัง Dev report
Coordinate PM/Dev/QA agents แบบ Pipeline Pattern — auto-route งานตาม role, ส่งต่ออัตโนมัติเมื่อ status เปลี่ยน, track progress ทุก handoff
Create color-coded teaching scripts / presenter talking-point documents (.md) that serve as a "live reading book" for speakers. The output is a markdown file where each color tells the presenter what to do: blue = speak this line, red = pronunciation guide, dark blue = slide header, black = stage direction & audience engagement, yellow box = terminology. Use this skill whenever the user mentions: teaching script, talking points, presenter script, key talking point, speaker notes, lecture script, training script, presentation script, สคริปต์การสอน, สคริปต์วิทยากร, คำพูดที่ต้องอ่าน, หนังสือสำหรับพูด, or any request to create a document that guides a speaker through a slide deck with exact words to say, pronunciation help, and stage directions. Also trigger when the user uploads a PowerPoint (.pptx) and asks to write a script, narration, or talking points for it.
Use for QA evidence, release readiness, rollback review, and close-out decisions.
Use for build completion evidence, code review readiness, and ready-for-QA checks.