메뉴
The Atelier deployment advisor — where to host frontend, backend, database, and workers, with real cost expectations. Load when choosing hosting, estimating costs, or setting up deploys/CI.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Atelier's native asset tools — image→webp compression, video→webp frame sequences, scrollytelling scaffolding, and AI video generation (Gemini Omni Flash / Veo 3.1) with a graceful no-credentials fallback. Load when handling images, video, storytelling sections, or media generation.
Turn any component site, asset library, or API into something agents can use — a registry entry, a skill, or a full MCP server. Load when the user says "add this site/library to Atelier" or wants to contribute a new integration.
The Atelier operating loop — understand → clarify → plan → execute → review → compare-vs-done-state. Governs when to stop, when to ask, which model tier runs each phase, and when to think deep or compact. Load for ANY nontrivial task.
The thinking protocol — load FIRST on every prompt, before any other skill. Makes any model (Opus, Sonnet, Haiku) reason like a principal engineer - decisive answers, root-cause debugging, security reflexes, outcome-first communication. This is step 0 of the loop.
The design-truth gateway — resolves which design files are authoritative right now (core law + self-learning overlays). Run at the START of any UI/design task, before reading the design law.
The Atelier Auth Law — production-hardened session architecture (rotating refresh tokens, families, grace window, silent refresh, sliding window) plus the anti-patterns that corrupt auth. Load before designing or touching ANY authentication code.
| name | deploy-advisor |
| description | The Atelier deployment advisor — where to host frontend, backend, database, and workers, with real cost expectations. Load when choosing hosting, estimating costs, or setting up deploys/CI. |
Recommend the cheapest option that meets the requirement; always tell the user the cost BEFORE they choose. Never pay for idle when scale-to-zero exists.
| Platform | Cost | Pick when |
|---|---|---|
| Cloudflare Pages/Workers | Free tier generous; paid from $5/mo (pay only when code runs) | Default for landing/static + edge SSR; unmetered bandwidth |
| Vercel | Hobby free; Pro $20/user/mo | Heavy Next.js features, preview deploys, team workflow |
| GitHub Pages | Free | Pure static docs/sites |
| Platform | Cost | Pick when |
|---|---|---|
| Google Cloud Run | Scale-to-zero; ~free small loads | Containerized API, spiky traffic |
| Cloudflare Workers/DO | $5/mo base; $0.15/M requests | Edge APIs, real-time DO, queues |
| Oracle Cloud free tier | $0 (4 ARM OCPU / 24GB — always free) | A real VPS for free; Celery/Redis/Postgres self-host |
| Hostinger / cheap VPS | ~$5–10/mo | Full control, predictable bill |
| AWS EC2 t3.micro / ECS Fargate | ~$8/mo+ / per-vCPU-hr | Team already on AWS |
| Cloudflare Tunnel | Free | Expose home/VPS backend with zero port-forwarding + DDoS protection |
| Need | Default | Cost |
|---|---|---|
| Postgres | Neon / Supabase free tier → managed when serious | $0 → ~$19–25/mo |
| Redis (cache/queues) | Upstash free tier / self-host on VPS | $0+ |
| Object storage | R2 (no egress fees) or S3 | R2 free 10GB |
| Video | Cloudflare Stream (HLS) | per-minute pricing |
| SES (cheapest at scale) / Mailgun to start | ~$0.10/1k |
